Energy Performance of

2%3)$%.4)!,
"5),$).'3
 EDITOR: M. Santamouris

Energy Performance of

2%3)$%.4)!,
"5),$).'3

! PRACTICAL GUIDE FOR ENERGY RATING AND EFFICIENCY

First published by James & James/Earthscan in the UK and USA in 2005

Copyright 
c M. Santamouris, 2005

All rights reserved

ISBN: 1-902916-49-2

Typesetting by S.R. Nova Pvt Ltd., Bangalore

Cover design by Paul Cooper Design

For a full list of publications please contact:

Earthscan
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada by Earthscan
711 Third Avenue, New Y ork, NY 10017
E arthscan is an imprint o f the Taylor &Francis G roup, an informa business

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

Energy performance of residential buildings : a practical guide for energy rating and
efficiency / edited by M. Santamouris.
p. cm.
Includes bibliographical references and index.
ISBN 1-902916-49-2 (hardback)
1. Dwellings-Energy conservation. I. Santamouris, M. (Matheos), 1956-
TJ163.5.D86E528 2005
697–dc22
2004013404
 v

Contents
PREFACE Measurements in climate chambers 16
Introduction on the Energy Rating of Buildings ix Hot box apparatus 16
M. Santamouris Outdoor test facilities 16
Why carry out energy rating of buildings? ix Detailed monitoring campaigns in non-occupied
Historical developments ix buildings 16
Problems and future prospects x General 16
The EUROCLASS method x Practical results 17
Description of this book – how to read the book xi Detailed monitoring in occupied dwellings 19
References xii Brief overview of interesting studies 19
Save HELP 19
CHAPTER 1 PhD thesis by Zoltan Somogyi 19
Review of selected theoretical and experimental Methodology 19
techniques for energy characterization of Error analysis 20
buildings 1 The impact of climatic region 21
P. Wouters and X. Loncour Normalized energy consumption 21
Introduction 1 Individual values for losses and gains 21
Northern cold climates 21
Part 1. State of the art – measurement techniques 1
Southern warm climates 21
Calculation methods and experimental techniques 1
Monitoring equipment 22
The university projects and the Energy Barometer 1
Temperature 22
The university projects 1
Solar radiation data 22
The task 2
Air change rate 22
Calculation model 2
Energy use 22
Background 4
Electricity and gas 23
Measurement model for energy consumption 4
Fuel, wood, etc. 23
The theoretical model for energy
Conclusions 23
consumption 5
References 23
The Energy Barometer 6
The university projects and the Energy Barometer 23
The calculation principles 6
Save HELP 23
The technical solution used to collect the data 7
STEM & PSTAR 23
The Save house energy labelling procedure (HELP) 8
Neural networks 24
Theoretical background 8
Other related topics 24
Determination of a test procedure 8
Other measurement techniques and monitoring
The choice of measured signals 8
experiences 24
The method of analysis 9
Calculation methods and parameter estimation 24
Experimental procedure 9
In situ evaluationof UA and gA values 25
STEM and PSTAR 10
Theoretical background 11
Applications of the method 12 CHAPTER 2
Neural networks 12 Experimental methods for the energy
Background theory 12 characterization of buildings 26
Applications of the neural network technique 13 Jan Akander and Gudni Jóhannesson
Supplementary examples of applications of the Definitions 26
neural network technique 14 The experimental protocols 26
Other related topics 14 The need for experimental protocols 27
Miscellaneous 14 The aim of the experimental protocol 27
Experimental techniques 14 The idealized model of energy flow in buildings –
Calculation methods and parameter estimation 14 definitions 28
Conclusions 15 Supplied energy 28
Delivered energy and specific energy 28
Part 2. In situ evaluation of UA and gA values – Energy for space heating 29
an overview of possibilities and difficulties 15 Energy for space cooling 29
Introduction 15 Energy for tap-water heating 29
Direct measurements of U-value 15 Energy for appliances and lighing 30
vi Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Energy for external appliances 30 Use of oil 48

Energy from metabolism 30 The use of natural gas 49
Energy from solar irradiation 30 Biomass/wood fuel, coke/coal 49
Climate-dependent and climate-independent District heating/cooling 49
variables 31 Tap hot-water measurements 49
Proposal of two protocols 31 Processing of data 49
Overview of BEP and MEP 31 The short monitoring period 50
Billed Energy Protocol (BEP) 32 The extended monitoring period 50
Strategy 32 Estimation of the heat-loss factor 50
Pros and cons 34 The energy signature method 50
Conditions for the application of BEP 34 The auditing phase 52
The requirements concerning the residence Development of the audit form 52
and the occupants 35 Recommendations for audits 53
The requirements concerning the billed and Auditing steps 53
recorded energy consumption 35 The audit form 53
The requirements concerning audit and General building information 54
climate information 35 Residents and behaviour 54
Step-by-step description of BEP 35 Heating 55
Pre-arrangements 35 Cooling 55
Collection of data on the supplied fuels 35 Domestic hot water 55
The audit 35 Appliances 56
Analysis of bills 36 Electricity meter information 56
Optional services 36 Suppliers 56
Results 37 References 56
Processing of data 37 Further reading 56
Frequency of bills 37
Tap hot water 37 CHAPTER 3
Supplied energy known 37 Energy normalization techniques 57
Delivered energy known 38 Jan Akander, Servando Alvarez and Gudni Jóhannesson
Supplied and delivered energies unknown 38 Introduction 57
Appliances and lighting 38 Space-heating energy 57
Stoves, fireplaces and other heating units Delivered space heating 57
without thermostat controls 38 Supplied space heating 57
External energy use 39 Space-cooling energy 57
External energy use within the envelope 40 Delivered space cooling 57
Delivered and supplied energy for seasons 40 Supplied space cooling 57
Delivered space heating 40 Global energy 57
Delivered space cooling 41 Delivered energy 57
Delivered daily base energy use 41 Supplied energy 58
The protocol form 42 Normalization of BEP and MEP output 58
Monitored energy protocol (MEP) 42 BEP 58
Strategy 42 MEP 58
Measurement strategy 42 Normalization with respect to the external
Procedure strategy 44 environment climate 59
Pros and cons 44 The heating and cooling degree-day method 59
Conditions for application of MEP 45 Limitations of the model 59
The requirements concerning the billed and Tolerances of the model 59
recorded energy consumption 45 The modified utilization factor (MUF) method 60
The requirements concerning audit and Introduction of the utilization factor for heat
climate information 46 gains 60
Specifications and requirements 46 Normalization of space heating energy 60
Climate gauge specification 46 Background information from the audit 61
Indoor temperature 46 Estimation of heat losses at the prevailing
Outdoor temperature 47 temperature 61
Solar radiation (optional for local Calculation of the modified gain-loss ratio 61
measurements) 47 Determining the potential heat losses at the
Humidity sensor (optional) 47 set-point temperature 61
Measurement of supplied energy 47 Heat losses for the reference year 61
Use of electricity 48 Normalization of space-heating energy 61
 Contents vii

Normalization of space-cooling energy 62 Estimation of the unknown parameters of the

Limitations of the model 62 energy consumption 79
Tolerances of the model 62 Energy consumption for hot water 79
A pre-study 62 Energy consumption for cooking 79
The Climate Severity Index 64 Normalization – degree-days correction 79
Determination of the CSI 64 The normal Belgian climate 80
Example and robustness of the CSI 65 Building no. 1 80
Use of the Climate Severity Index 66 General information 80
Normalization with respect to the internal Available data – calculation and results 81
environment climate 67 Building no. 2 81
The heating and cooling degree-day method 67 General information 81
The modified utilization factor method 68 Available data – calculation and results 82
Normalization with heat-loss factor methods 68 Swedish buildings 85
The Energy Signature method 68 The Swedish billing system 85
The UA and gA method 69 Climate data 85
Normalization with respect to building size 69 Building no. 1 86
Definition of the floor area 69 General information 86
Area-specific space-heating energy 69 Available data – calculation and results 87
Equivalent area-specific space-cooling energy 69 Remarks 88
Area-specific global energy 70 Building no. 2 89
References 70 General information 89
Available data – calculation and results 92
CHAPTER 4 Remarks 92
The Euroclass method – description of the Spanish buildings 94
software 71 Introduction 94
Servando Álvarez, Antonio Blanco, Juan Antonio Sanz and Building no. 1 96
Francisco J. Sánchez General information 96
The rating methology 71 Available data – calculation and results 96
Description of the software 71 Remarks 96
Input and output sheets 71 Building no. 2 97
The database 72 General information 97
Climatic data sets 72 Available data – calculation and results 97
Inertia ratios 72 Remarks 97
Heating reference conditions 72 Greek buildings 97
Cooling reference conditions 72 Introduction 97
Reference rating distribution 72 Building no. 1 99
Distribution calculation set of values for General information 99
references 72 Available data – calculation and results 101
The calculations sheet 72 Remarks 101
Input and output screens 72 Building no. 2 101
Description of variables 72 General information 101
Description of the input 72 Available data – calculation and results 103
Description of the output 72 Remarks 103
Description of the database 72 Comments on the results for Greek buildings
Example 72 1 and 2 104
Conclusions concerning the BEP procedure 104
CHAPTER 5 Results for the Monitored Energy Protocol (MEP)
Examples and case studies 78 procedure 106
P. Wouters and X. Loncour Introduction 106
Introduction 78 Belgian buildings 106
Building selection – characteristics of the building Building no. 1 106
analysed 78 The monitoring carried out 106
Selection criteria 78 Available data – calculations and results 106
The type of buildings selected in the four countries 78 Building no. 2 107
Procedures and normalization techniques applied 78 The monitoring carried out 107
Results of the Billed Energy Protocol procedure 79 Available data – calculations and results 107
Introduction 79 Swedish buildings 109
Belgian buildings 79 Building no. 1 109
The Belgian billing system 79 The monitoring carried out 109
viii Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Available data – calculations and results 109 The monitoring carried out 118
Analysis of wood consumption 110 Available data – calculation and results 118
Analysis of energy for tap-water heating 111 Remarks 119
The heat-loss factor 111 Comments on BEP and MEP results for the two
Remarks 113 Greek buildings 119
Building no. 2 114 Conclusions concerning the MEP procedure –
The monitoring carried out 114 comparison with BEP 120
Available data – calculation and results 114 References 123
Remarks 114
Spanish buildings 114 APPENDIX 1
Building no. 1 114 Audit form 124
The monitoring carried out 114
APPENDIX 2
Available data – calculation and results 115
Energy transmittance by glazing and shading
Greek buildings 116
factors 132
Introduction 116
Building no. 1 116 APPENDIX 3
The monitoring carried out 116 Estimated average fuel combustion efficiency of
Available data – calculation and results 116 common heating appliances 134
Remarks 116
Building no. 2 118 INDEX 135
 ix

PREFACE
Introduction on the Energy Rating
of Buildings
M. SANTAMOURIS
Group Building Environmental Studies, Physics Department, University of Athens, Greece

WHY CARRY OUT ENERGY RATING Following the energy crisis, in 1974 the Swedish
OF BUILDINGS? government introduced financial support in the form of
loans and subsidies for energy-saving measures within the
Energy efficiency is a critical issue for high-quality housing.
building stock. The aim was to stimulate efficient energy
Energy not only represents a high percentage of the running
use and to reduce energy for heating. The specific goal was
cost of a building but it also has a major effect on the thermal
to decrease the gross heating consumption for residential
and optical comfort of the occupants.
areas by 39–48 TWh over a 10-year period (1978–1988) with
Recent developments in energy technology make it
a total investment of 31–48 billion SEK (1977 value). The
possible to decrease significantly the energy consumption
retrofitting measures were voluntary.
of buildings, to create housing that is more comfortable
In order to evaluate the plan during the initial stages,
and to implement a major decrease in emissions to the
a programme was set up during the first three years
environment.
(1977–1980) to assess the characteristics of the building
Although the energy efficiency of many home products,
stock and to estimate energy savings. The aim was, on
(dishwaters, refrigerators, etc.), is available to consumers,
the basis of collected data on energy bills and building
the relative energy consumption and efficiency of one
technical descriptions etc., to estimate the actual mean
of the most expensive investments, their dwelling, which
energy savings due to several retrofitting methods. In total,
generates the highest energy bills, is not available to the
1,144 buildings and apartments were audited, of which 944
consumer.
were single-family houses and 200 were multi-family
Energy rating of a dwelling can provide specific
buildings.1
information on the energy consumption and the relative
The continuation of the project2 focused primarily
energy efficiency of the building. Energy rating is per-
on experimental and theoretical evaluation of energy
formed through standard measurements carried out
conservation measures on statistically selected objects within
under a specific experimental protocol by specialized and
the Swedish building stock. The results were based on
accredited professionals. It is then possible for a potential
measurements before and after retrofitting during the
buyer to have exact information on the energy bills he has
period 1982–1986 in approximately 300 single- and multi-
to pay, while the owner of a house may be able to identify
family houses in seven municipalities.
and pinpoint specific cost-effective improvements.
The Energy Barometer (EB) idea3 was then developed
Energy audits involve specific measurements of the
in Sweden and concerns monitoring the development of
building shell, such as insulation levels and window
building energy use via continuous energy and climatic
efficiency, of the lighting and the ventilation, as well as
measurements in houses and reporting changes of building
of the heating and cooling systems of the building. The
energy over a short time span. The information system was
behaviour of the occupants, who explicitly control and affect
made up of two parts:
the internal environment, is also considered. The results are
normalized and the building is given a score, for example
between 1 and 100, that makes it possible to classify the • One was aimed at providing for the wider public
building against an absolute performance scale. information on the actual and the predicted energy use of
the dwellings, together with analyses of trends and effects
of energy-related measures on a large scale. The estimates
are based on a representative statistical sample from the
HISTORICAL DEVELOPMENTS population of interest. A sample size of around 1,000
Energy rating started just after the energy crisis. The buildings enabled statistically reliable monitoring of how
concern of the industrialized countries about the high the Swedish national use of energy has varied over time.
energy consumption of the building sector and, in particular, • The second part aimed mainly at providing individual
of residential buildings initiated actions and programmes house owners with a means of monitoring their energy
aiming to rationalize the energy consumption of dwellings. bills. Occupants of a building connected to the Energy
x Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Barometer could analyse their own energy situation and PROBLEMS AND FUTURE PROSPECTS
also view it in relation to that of the population as a whole.
Energy rating of dwellings has reached a high level of
scientific maturity, in particular in countries where energy
In the USA, energy rating systems have been used since rating has been used for many years. It is expected that
the 1980s. However, the idea of using Energy Efficiency the new European Directive for Energy Efficient Buildings
Mortgages (EEM) associated with home energy rating will enforce the use of such methodologies and will expand
systems, which has been applied since 1990, has considerably the application of home rating systems in all European
increased the penetration of rating systems in the residential countries and, in particular, in Southern Europe. The
market. The existence of an energy audit has substantially new Directive is asking for an energy rating of many
helped the mortgage industry to make loans for energy types of buildings and brings energy certification into
improvements. In 1992, the US Congress voted through the everyday life. Member States have to prepare their national
Energy Policy Act, which provided for the establishment methodologies, but an effort is made to homogenize the
of credible voluntary national guidelines for home energy rating techniques to be used as much as possible.
rating schemes.4 Then a Home Energy Rating System However, a number of barriers to the widespread use
(HERS) program was started in five pilot states, Alaska, of home rating systems have been identified in many
Arkansas, California, Vermont and Virginia. Today 31 states countries. The main barriers appear to include:
have adopted the MEC 1993 version or its equivalent.
However, only 2% of new homes actually receive an energy • lack of owners’ awareness of energy efficiency benefits
rating in the USA, and most are utility programmes with • insufficient awareness and training of property managers,
tax-payer subsidies. builders and engineers
In Denmark, energy rating schemes have been prepared • low energy costs for both electricity and thermal energy
for large commercial buildings since 1992. The schemes • lack of sufficient funding to assist the penetration of
were extended to residential buildings, a year later, 1993. home rating systems in the real market
The scheme is mandatory, relatively costly and quite • the relatively high cost of home energy systems
comprehensive. • lack of data on the energy consumption of dwellings, at
In the UK, in the 1980s and 1990s Building Research least in many European countries, which do not permit
Establishment (BRE) performed hundreds of multi-year comparisons to be made and defaults defined
energy audits in residential buildings. From these, BRE was • lack of specialized professionals to perform energy audits
able to develop the Domestic Energy Model (BREDEM). and ratings in residential buildings
Today three labelling schemes are in use: • lack of builder incentives
• lack of financial interest and lack of financial gains for
the owners, the builders and the real-estate managers.
• the Standard Assessment Procedure (SAP) on a scale
from 1 to 120, which has been required by building During the last few years, environmental issues other
regulations for new housing since 1995 than energy have been considered in the assessment of
• the National Home Energy Rating, (NHER), on a scale buildings.5 Ecological parameters such as shortage of raw
of 0 to 10 materials, water consumption, indoor air quality, noise and
• an estimate of the carbon dioxide emitted each year as a pollution, health aspects and waste treatment are the main
result of a home’s energy use (the carbon index). considerations. It is believed that environmental rating
systems, involving energy issues and life-cycle analysis, will
The principal target for labelling in the UK is social housing have a very fast development in the near future.
and today almost 75% of the social rental housing has been
labelled, while about 20% of owner-occupied housing has
been labelled at the time of sale. THE EUROCLASS METHOD
In Ireland, the National Irish Centre for Energy Rating As described above, several energy rating techniques have
(NICER) created the Energy Rating Bench Mark (ERBM) been proposed and are in use. Each methodology has
in 1992 to deal with existing buildings. Almost 8,000 new to be based on an experimental protocol for collecting
houses per year are labelled with ERBM. energy data, a theoretical algorithm to normalize the energy
In Spain, and in particular in the Basque Country, consumption and an algorithm to classify buildings. It is
an energy rating system has been developed to classify thus very reasonable for each national methodology to
new residential buildings. Energy-efficiency certificates are be adapted to the characteristics of the national building
awarded to buildings in two stages. A first certificate is stock, the national methodology for measuring energy
awarded in the design phase and a final one is provided to and the specific climatic characteristics of the country.
the finished building. For example:
Finally, in the Netherlands, a rating scheme, EPB, was
developed in the mid 1990s. The method mainly targets • Cooling has only very recently become important for the
social housing. A new method, EPA, was put into operation residential sector. Thus, most of the existing rating
during the year 2000. methodologies do not consider at all the specific energy
 Preface xi

consumption for air conditioning. Given the general the frame of the European SAVE program, EUROCLASS.6
energy tendency in Europe and the requests of the new The project was coordinated by the Group of Building
European Directive, a new European classification Environmental Studies of the University of Athens,
methodology must involve a cooling methodology. Greece.
• In Nordic countries solar gains do not significantly The overall book is made up of five chapters plus three
influence the heat balance of the building and thus a Appendices. The methodology has been prepared with the
normalization approach based on degree days seems very active participation of all partners in all parts of the work.
reasonable. However, such an approach can be used in The authors mentioned in each chapter have undertaken
Southern countries, where solar and internal gains may to report the specific work carried out by the whole
play a very important role. consortium.
• The characteristics of residential buildings differ Chapter 1 deals with the presentation of some selected
considerably from state to state. Different types of theoretical and experimental methodologies proposed
materials are used and different construction and applied for the energy characterization and rating
techniques are employed, while the mode of of dwellings.
operation is sometimes completely different. Thus, the Chapter 2 deals with the presentation of the pro-
benefit of a utilization factor is doubtful as long as posed experimental protocols – methods for the energy
generalization of data from numerous buildings in a state characterization of dwellings. Two separate protocols have
makes the results for one specific building uncertain. At been designed to fulfil the same requirements, but with
the same time it does not allow the use of the method in different degrees of detail. The costs, the resources, the
another state. delivered information (service) and the level of accuracy of
• Billing of the energy consumption differs between the the results will allow the customer (occupants) to choose
various countries. A methodology to calculate the specific which protocol to use. The results generated by each
energy consumption of a building based on energy bills is protocol are compatible in rating since both have a similar
highly influenced by the billing methodology and thus experimental platform.
national methodologies need to be fully adapted to the The two protocols are called:
specific conditions.
• The energy consumption of the residential sector varies • Billed Energy Protocol (BEP)
considerably between states. Classification of buildings • Monitored Energy Protocol (MEP).
based on their energy consumption is always based on a
large database of national energy data and thus the
Chapter 3 deals with normalization techniques. In a general
divisions between the classes have a very strict national
way, normalization takes into account:
character.

• the size of the building

EUROCLASS is a new methodology recently developed • the climate of the external environment
through the European SAVE program and aims to over- • the climate of the internal environment.
come most of the above problems. EUROCLASS has run
between 1998–2000. The method proposes experimental The main existing normalization techniques for space heat-
protocols that may be used in and adapted to each state, ing and space cooling are presented and their limitations
while it also proposes a theoretical method that includes and tolerances are discussed in detail. A new proposed
all specific energy uses and treats energy normalization in normalization method, the Climate Severity Index, is
a very innovative way. The proposed methodology seems to then presented. The method presents many advantages
provide answers to most of the problems of the existing compared with the classical normalization techniques.
methodologies and thus may be the platform for future Chapter 4 describes the software developed within the
developments in the field. frame of the EUROCLASS project, which is designed
to apply the proposed rating methodology for dwellings.
EUROTARGET, which is the name of the software, is
described in detail and examples are given. A full copy of
DESCRIPTION OF THIS BOOK – HOW the software is enclosed with the book.
TO READ THE BOOK The methodology presented in the above chapters has
The aim of this book is to inform readers about the been applied in four countries (Belgium, Greece, Spain and
latest developments in the field of energy rating of dwel- Sweden). Chapter 5 describes the results of the procedure,
lings. In particular, the book presents the new European the problems encountered and the solutions envisaged to
experimental and theoretical methodology for assessing and solve these problems.
rating the energy efficiency and performance of dwellings. Chapter 5 is divided into two parts. The first part deals
The book is a collaborative work undertaken by the with the results of the BEP procedure, while the second part
University of Athens, Greece, the Belgium Building deals with the results of the MEP.
Research Institute (BBRI), the Royal Technical University The two protocols have each been applied in the four
of Stockholm and the University of Seville in Spain, in considered countries. The application of the methodology,
xii Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

the way to carry out the monitoring, the choice of 2. Elmroth A, Hjalmarsson C, Norlén U, Rolén C, et al., 1989,
normalization techniques and the calculations have been Effekter av energisparåtgärder i bostadshus, Rapport R107:1989,
implemented at a national level. Byggforskningsrådet, Stockholm, Sweden.
Finally, Appendix 1 presents the audit forms prepared 3. Norlen U, 1985, ‘Monitoring energy consumption in the Swedish
building stock’, Proceedings of Conference on Optimisation of Heating
and used in the frame of the Euroclass project. The forms
Consumption, Prague, Swedish Institute for Building Research,
should be very useful for anyone interested in applying the
Gävle.
proposed rating methodology. 4. Fahrar B C, 2000, Pilot Program Report: Home Energy Rating Systems
and Energy – Efficient Mortgages, NREL/TP–550-27722. Available
through NREL.
5. Environmental Assessment of Buildings, 1995, A Thermie Pro-
REFERENCES
gramme Action, No B 108, European Commission, Directorate
1. Bostadsdepartementet, 1980, Energispareffekter i Bostadshus där General for Energy, Brussels.
Åtgärder Genomförts med Statligt Energisparstöd. Expertbilaga 5 6. Santamouris M (editor), 2001, Final Report of the EUROCLASS
till SOU 1980:43 – Program för energihushållning i befintlig Project, SAVE program, European Commission, Directorate
bebyggelse. Ds Bo 1980:8. General for Transport and Energy, Brussels.
 1

CHAPTER 1
Review of selected theoretical and
experimental techniques for energy
characterization of buildings
P. WOUTERS AND X. LONCOUR
Division of Building Physics and Indoor Climate, Belgian Building Research Institute, Brussels

INTRODUCTION measuring techniques (for instance, the use of the

Internet) and are therefore less interesting within the
The aim of this chapter is to present selected theoretical
context of this project.
and experimental energy rating methodologies developed
for dwellings. It is not the intention of the authors to present
We mention in this chapter the different calculation
all existing methodologies, but to inform the readers about
methods. We give the references of the papers concerned
the main types of proposed theoretical methodologies and
and, when they are available, we describe the techniques
to present representative methods of each type.
used to collect the relevant information on site.
This chapter consists of two parts:
The different calculation methods that are considered in
this document can be regrouped into five categories:
• The first part presents the result of the bibliographic
study on the state of the art of the theoretical and 1. the university projects and the Energy Barometer
experimental techniques used to determine the energy 2. the Save HELP method
consumption of residential buildings. 3. short-term energy monitoring and primary and
• The second part deals with the in situ identification of the secondary term renormalization
transmission heat loss coefficient + (UA value) and the gA 4. neural networks
values (characterizing the solar gains) of buildings. 5. other related methods.

The university projects and the

Part 1. State of the art – Energy Barometer
measurement techniques THE UNIVERSITY PROJECTS

CALCULATION METHODS AND EXPERIMENTAL The two ‘university projects’ have been developed in Sweden.
TECHNIQUES References [1] and [2] describe these two projects. Since
these references are in Swedish, we give more detail here
Several remarks need to be made before going into more about the methodology applied in the scope of these two
detail: projects than for other methods described in this document.
In 1974, the Swedish government introduced financial
• Much information can be found here about the support in the form of loans and subsidies for energy-saving
calculation methods used to determine the energy measures within the existing building stock. The aim was to
consumption of the buildings. stimulate efficient energy use and to reduce energy for heat-
• Some of these calculation methods are firmly based on ing. The goal was to decrease the gross heating consumption
experimental data. It should be noted that very often the for residential areas by 39–48 TWh over a 10-year period
articles focus more on the description of the calculation (1978–1988) with a total investment of SEK 31–48 billion
method than on the way of realizing the experimental (1977 value). The retrofitting measures were voluntary.
monitoring. In order to evaluate the plan during its initial stages, a
• In general, few papers focus on the experimental programme was set up during the first three years (1977–
techniques used to collect the data on site. 1980) to determine whether or not it was worth continuing
• Some of the experimental techniques described in the the plan. The evaluation was also intended to assess the
literature are quite old (some of them were developed characteristics of the building stock and to estimate energy
more than 20 years ago). These experimental techniques savings. The aim was to estimate, on the basis of collected
do not take into account the latest developments in data on energy bills and building technical descriptions, the
2 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 1.1 The efficiency of heating systems during winter and summer. Values are given for the various types of single- and multi-family dwellings
Efficiency Electrical heating Oil/gas District
Single-family Multi-family Single-family Multi-family Single-family Multi-family
ηwinter 1.00 1.00 0.65 0.85 0.95 0.95
ηsummer 1.00 1.00 0.30 0.65 0.95 0.95

actual mean energy savings resulting from several retrofitting Table 1.2 Fuel conversion factors
methods. Reference [1] describes this project, which is called Fuel (unit) Conversion factor
the ‘university project’ (UP1), and the method applied. Oil (m3 ) 1.000
Gas (m3 ) 0.471
The task Electricity (kWh) 0.101
District heating (kWh) 0.101
The task was to calculate the energy savings resulting from
different retrofitting means on the basis of collected energy-
bill data and of an inspection of building technology and the Table 1.3 Values assumed for energy consumption by appliances and
systems installed. In total, 1,144 buildings and apartments hot water
were audited, of which 944 were single-family houses and Energy (kWh) Single-family house Multi-family house
200 were multi-family buildings. Broadly, the following steps Electrical appliances 4,600 2,800
were taken: Hot water 4,000 3,500

• Energy bills were collected before and after the

retrofitting was carried out. Energy for appliances and hot water
• Bill values were normalized according to temperature The use of energy for electrical appliances and hot water was
variations over the years. assumed to be as shown in Table 1.3 for each household.
• The energy saved was measured in corresponding litres
of oil, during a reference year. Degree-hours
Among the climatic factors, such as temperature, solar
It was found that 841 buildings had a complete set of bills
irradiation, wind, snow, long-wave irradiation and moisture,
and 303 had an incomplete set. More details are given below,
that influence the heat balance of a building, the outdoor
especially on the underlying assumptions made.
temperature was considered to be the most important
The static energy signature was determined for each
factor. Because of the seasonal variations in outdoor
building, and from this data could be statistically applied
temperature, the periods taken into consideration before
to the Swedish building stock as a whole.
and after retrofitting were chosen so as to be complete years,
if possible. The number of degree-hours was computed from
The indoor temperature
The indoor temperature was assumed to be 21◦ C, with Q = T(θi − θe ) (◦ C · h)
the exception of the period during and shortly after the where T is the number of hours during the heating season,
oil crisis. The motivation for this was that information on θi is the mean indoor temperature and θe is the mean
indoor temperature that was supplied by the inhabitants outdoor temperature based on mean monthly values. Linear
was considered to be inaccurate. interpolation was used to calculate the mean outdoor
temperature.
The outdoor temperature For the calculations, the heating season was limited to
The Swedish Meteorological and Hydrological Institute months with sufficiently low outdoor temperature by starting
(SMHI) supplied outdoor temperatures on the basis of data in October and ending in April.
from climate stations across the country. Surveyors had to In order to compute the number of degree hours for a
judge which climate station best described the temperature reference year, the collected outdoor temperatures for the
at a particular site. This eliminated local variations of years 1972–1979 were used.
temperatures.
U values
The efficiency of the heating system U values were calculated in accordance with norms. For
The efficiency of heating systems varied from building to double- and triple-glazed windows, the values were 3.0 and
building, but within this work was assumed to be the same 2.0 W/(m2 · ◦ C), respectively. The glazed part was assumed
for various categories of houses. The values decided upon to be 70% of the gross window area.
are listed in Table 1.1.
Energy saving calculations were made as if all buildings
Calculation model
were heated with oil and a conversion factor was used for
each fuel used as a heat source, based on the energy content The calculation model was based on an energy balance for
of the fuel (Table 1.2). the building, which was split up into a summer season and a
 Review of selected theoretical and experimental techniques for energy characterization of buildings 3

c = the winter factor, litres of oil per hour

d = the summer factor, litres of oil per hour

The two first terms on the right-hand side of this equation

describe the energy consumption during winter, while the
last is for the summer period. The heat-loss factor b is a
measure of the insulation and of the amount of ventilation
of the building. The summer factor d describes the energy
consumption during the non-heating season (summer),
whereas the winter factor c is more difficult to interpret. All
three factors depend on the efficiency of the heating system.
The factors are determined by means of linear regression on
data from buildings with extensive information on energy
Figure 1.1 Net energy balance for a building. The shaded parts of consumption before and after retrofitting.
the net supplied energy depict the energy supplied for an electrically With the assumption that the winter factor c and the
heated building summer factor d are known and remain unchanged during
the period before and after retrofitting, the heat-loss
winter season. Figure 1.1 illustrates the various energy parts factors before retrofitting, b b , and after retrofitting, b a , are
that make up the whole. 1
bb = (W b · H − c · Tb − d · Tb )
In Figure 1.1 and the equations below, the following Qb
abbreviations are used:
and
SG = solar gains 1
ba = (W a · H − c · Ta − d · Ta )
PG = heat from people Qa
EG = heat from household appliances (electrical gains)
HW = net heat supplied for hot water Now all these values can be determined and can be used
HS = net heat supplied by the heating system to calculate the energy consumption that would have been
TL = heat loss due to transmission measured during a year with ‘normal’ climate:
VL = heat loss due to ventilation (W · H )b,r = bb · Q r + c · Tr + d · Pr
DL = heat loss via drainage
and
For electrically heated buildings, the energy consumption is
(W · H )a,r = ba · Q r + c · Tr + d · Pr
W = TLw + VLw + DLw − PGw − SGw + HWs + EGs
where index r denotes the reference year.
For oil-, gas- or district-heated buildings, the energy
The saving of energy due to retrofitting is the difference
consumption is
between (W · H )b,r and (W · H )a,r . First, the net energy has
W = (TLw + VLw + DLw − PGw − SGw − EGw )/ηw to be calculated with the efficiency of the actual heating
system. Then, divided by the efficiency of an oil-based boiler,
+ HWs /ηs
the saving expressed in litres of oil per year is found to be
If it is assumed that the losses due to transmission and
ηw
ventilation are proportional to the amount of degree-days, S = (bb − ba )Q r
ηw oil
that the drainage loss, the heat gains from people and solar
irradiation and household appliances are proportional to the where
number of hours in the heating season (winter) and that the S = the energy saving if the building were heated using
household appliance gains during the non-heating season oil (litres/year)
are proportional to the number of hours during the non- bb = the heat-loss factor before retrofitting
heating season, the two equations can be rewritten in the (litres/degree per hour)
form ba = the heat-loss factor after retrofitting
W ·H = b ·Q +c ·T +d · P (litres/degree per hour)
Q r = the number of degree-hours during a reference
where year
W = the collected energy consumption ηw = the efficiency of the actual heating system
H = conversion factor during winter
Q = the number of degree-hours during the period ηw oil = the efficiency of the oil-based heating system
considered
T = the number of hours during the heating season The second ‘university project’ (UP2) described in
P = the number of hours during the non-heating season reference [2] focuses primarily on the experimental and
b = the building heat-loss factor, litres of oil per hour theoretical evaluation of energy conservation measures on
per degree statistically selected objects within the Swedish building
4 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 1.4 Correspondence between measured and theoretically performance parameters. For both models, the following
predicted savings data are required:
Saving in relation to energy
• parameters that describe the energy performance of
Retrofit Building type consumption prior to retrofit
the building
Measured (%) Predicted (%)
• the indoor temperature
Triple glazing Single-family 6 10
• climatic parameters that describe the climate of the
Retrofit package Single-family 19 25
normalized year.
Conversion to Single-family 22 20
electric heating The energy saving is the difference between the energy
Triple glazing Multi-family 9 10 consumption before and after retrofit. The energy saving
Additional insulation Multi-family 5 6 from the measurement model is called the measured energy
of attic saving, whereas the value calculated from the theoretical
Regulating package Multi-family 4 3 model is called the expected energy saving. The energy
Retrofit package Multi-family 14 16 saving is dependent upon which indoor temperature is used.
District heating Multi-family 24 15 A way of defining the energy saving is as follows:

[energy saving] = [effect of retrofit]

 
stock. The results were based on measurements before
effect of changed
and after retrofitting during the period 1982–1986 in +
indoor temperature
approximately 300 single- and multi-family houses in seven
 
municipalities. The most important results were: effect of other
+
influencing factors
• In most cases, energy conservation measures gave
The analysis was set up so as to minimize the last term (i.e.
statistically established savings. In many cases, however,
to make it negligible). If this is done, the following applies:
savings were small.
• For most cases, the measured savings agreed with the • The measured energy saving: any change in indoor
theoretically predicted savings (Table 1.4). temperature is represented in this change such that
• Energy savings of heat pump installations in single-family  
houses were 40–60% of pre-retrofit energy consumption. energy saving in
= [effect of retrofit]
• The temperature decreases expected as a result of some measured temperature
retrofits were not achieved.  
effect of changed
+
indoor temperature
Background
• The expected energy saving: the indoor temperature
UP2 was a continuation of UP1, but one of the aims was to before and after retrofitting has standardized values
improve the accuracy of the results where the procedures that are assumed to give the same thermal comfort.
of UP1 were considered to be weak. Monitoring indoor • Other types of energy savings: with the two models
and outdoor temperatures led to improvements, as did the (measurement and theoretical), and two temperatures
use of energy meters, which were read on a weekly basis. (measured and standardized), four different energy
In addition, oil burners were supplied with run-time sensors. saving variants can be calculated for every building.
Two models were used in the analysis: the measurement The measured energy savings (with measured indoor
model and the theoretical model. Common to both is that temperature) and the expected savings (with
the most important parameters of the energy performance standardized indoor temperature) are those
of the building are considered. that are mostly used in this report.
Energy consumption is presented in kWh per apartment
per normal year (here, energy consumption is that required
for space and domestic water heating during one year). The
Measurement model for energy consumption
climate is normalized to a year where: Calculations are made according to
W = b · Q + c · T + d · P (kWh/apartment/year)
• the outdoor temperature is equal to the mean where
temperature during the years 1951–1979 W = the annual energy consumption per apartment per
• the solar irradiation is equal to the mean solar year; in certain cases this is converted to kWh by
irradiation during the years 1955–1979 means of conversion factors that correspond to the
energy content of the fuel (kWh/apartment/year)
In the measurement model, series of weekly collected energy Q = the number of degree hours during the heating
consumption data were used. In the theoretical model, season (◦ C · h/year)
the descriptive technical information was collected and T = the number of hours during the heating season
used, together with default values, to determine the energy (h/year)
 Review of selected theoretical and experimental techniques for energy characterization of buildings 5

consumption during the periods when measurements

were made.
From 50 climate stations across the country, daily mean
1 b .Δ + c outdoor temperatures were obtained. For a particular
building, there is often a systematic difference between
the outdoor temperature at the building and the chosen
climate station, usually the nearest to the building site.
A correction was made for this difference between the
measured mean temperature at the building and that at the
climate station during the measurement period.
d

The theoretical model for energy consumption

0 If an energy balance model with access to outdoor climate
20 Temp diff Δ (°C)
data is set up, the annual energy consumption can be
calculated before and after a retrofit. The calculated energy
Figure 1.2 The relationship between the energy consumption per
consumption for space and water heating is found by
apartment per unit of time and the difference between indoor and outdoor
setting the supplied energy equal to the energy loss. The
temperatures, 
time period of the heating season is set to be the same as in
the measured case.
P = the number of hours during the summer season Certain assumptions were made concerning the energy
(h/year) for household appliances, the energy for heating water,
b = the building heat-loss factor the energy from persons and the gains from insolation
(kWh/apartment/◦ C · h) (see UP1 and below).
c = the winter factor, which quantifies temperature- The supplied net energy (i.e. the paid-for energy) for
independent energy losses (ground, drainage) heating space and water can be formulated as (see above
and heat gains (household appliances, persons, for meaning of the notation)
insolation) (kWh/apartment · h)
d = the summer factor, which quantifies the mean Wnet = HSw + HWw + HWs = (TLw + VLw )
energy for heating of water during summer + (DLw − EGw − PGw ) − SG + HWs
(kWh/apartment/h)
which is rewritten as
Figure 1.2 shows the relationship between the energy
consumption per apartment per unit of time and the Wnet = b(θi − θe ) · T · ηw + c · T · ηw − f · I · T · ηw
difference between indoor and outdoor temperatures, . + d · P · ηs
In the model, values for b and c are estimated from
measured values by the least-squares method, such that with the introduction of two new parameters:

n
f = the solar area, which is a fictitious window that
Wi (i − )
i =1 transmits all insolation (m2 /apartment)
b= and c = W −b · 

n I = the mean solar irradiation during the heating
(i − )2
i =1 season (kWh/m2 /h/year)
where
The following assumptions have been made:
Wi = the weekly mean energy consumption during
period number i divided by the length of the
• The sum of the transmission and ventilation losses is
period in hours
proportional to the number of degree hours.
i = the mean temperature difference (indoor −
• Drainage losses, the heat from household appliances and
outdoor) during period i
the heat from people are proportional to the length of
W and  are the mean values for the whole measurement the heating season.
period in question. The summer factor d has the same • The energy for water heating during the warm season is
standardized value for houses of the same type. proportional to the length of the season.
Alternative measurement models have been analysed,
for example one with a third energy parameter that The gross energy consumption is therefore
represents insolation. No major differences in the results
Wgross = [b(θi − θe ) + c − f · I ] · T + d · P
for the heating season were noted between the three-
parameter model and the two-parameter model. Some The heat-loss factor b is calculated as the sum of
small differences were explained by correlation between the transmission loss factor for the envelope components (the
week-wise solar irradiation and the outdoor temperature and U · A value), and the ventilation loss factor (0.33 · n · V
by the slight influence of the solar irradiation on energy with n representing the rate of air change in volume V).
6 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

The loss factor can therefore be expressed as period and let I be the global solar radiation on a horizontal
surface during the heating period (kWh/m2 ). Then the
U · A + 0.33 · n · V
b = 0.001 (kW/apartment/K) following expression can be used to calculate the annual
ηw
energy use, E i (kWh), for the heating in house i:
The winter factor c and the summer factor d are given default
Ei = ci Ti + bi Ti θ + fi Ii + di Pi
values.
The solar w aperture is calculated with four reduction The energy parameters c, b, f and d have specific values
factors and the gross window area A w , such that for each house. The climatic variable variables T, P and I
fs · f f · f c · f t depend on the climatic conditions for the region where
f = Aw (m2 ) the house is situated. The number of degree-hours, Q =
ηw
Tθ, differs between houses depending on the temperature
with conditions and the length of the heating season. The
fs = the shading factor variables T, P and I may to a first approximation be given
ff = the frame factor the same values for all houses in a particular region.
fc = the curtain factor For the purpose of estimating total energy for heating,
ft = the transmittance factor it is sufficient to include in the model solar irradiation on
a horizontal surface.6 When the end points of the heating
The Energy Barometer season are known, the contribution from the solar irradiation
during the heating season of a normal year can be estimated
The idea of the Energy Barometer (EB) is to monitor the using a sinusoidal approximation.
development of building energy use via continuous energy The energy consumption is determined on the basis of
and climatic measurements in houses and to report changes a static energy balance. The purpose is to estimate the
of building energy over a short time span. Analyses will ‘normal’ annual energy use for heating (space and tap
be presented on line. The word ‘barometer’ is intended water). Under stationary conditions, heat will be stored in
to refer to measuring the pressure on the energy market. or released from the house over a specific period of time.
The information system is made up of two parts. The first This assumption enables the formulation of a linear static
part is aimed at the population level, providing estimates of heat balance equation for the house – these are called energy
actual and predicted energy use, together with analyses of signature models.7
trends and the effects of energy-related measures on a large By dividing the previous equation by T, we obtain the
scale. The estimates are based on a representative statistical average energy use per unit time w (kWh/h) during
sample from the population of interest. A sample size of the heating season. The constant c is the average energy use
around 1,000 buildings makes possible a statistically reliable per unit time when there is no temperature difference and
monitoring of how the Swedish national use of energy varies no solar irradiation. ε is the error term:
across time.
The second part is aimed at providing individual house w = c + dθ + fs + ε
owners with a means of monitoring their energy cost
budgets. Buildings connected to the EB will not only be able The model assumptions are valid provided that the time
to analyse their own energy situation but also to see it in period chosen is long enough for the heat stored in or
relation to the population as a whole. released from the building to be very small compared to the
The Energy Barometer has been developed to measure total energy supplied during the period.
energy use in relation to internal and external climates in Two reasons for using a dynamic energy balance model
different types of single-family house during periods with instead of a static model of the type discussed above are:
and without heating. It is also expected to provide a basis
for analysis and evaluation of energy efficiency measures. • Energy use in buildings is a dynamic process, e.g. the
The electrical energy use by household appliances and the thermal inertia delays the ‘response’ of the building to
energy use for heating, including heating of tap water and changes in the outdoor climate.
indoor temperature, are measured in each selected building. • Information in the data is lost in the aggregation from
The whole theoretical background is described in references the hourly data to the daily or weekly data needed for
[3–5]. the static model.

The calculation principles A general dynamic model can be written as an equation that
expresses the energy use w(t) during hour number t as a
Two kinds of model can be used, based on either a static or
linear function of the following variables at hour t and the
a dynamic energy balance. The background relating to the
preceding hours (t − 1), (t − 2), . . .
static heating balance continues here.
Consider a whole year consisting of T hours when the
house is heated (‘winter hours’) and P hours when the house • the energy use w(t − 1), w(t − 2), . . .
is not heated (‘summer hours’). Let θ be the average • the indoor temperature θ in and the indoor temperature
indoor–outdoor temperature difference (◦ C) during this θ in (t − 1), θ in (t − 2), . . .
 Review of selected theoretical and experimental techniques for energy characterization of buildings 7

• the outdoor temperature θ out and the outdoor A first application of the method has been realized,
temperature θ out (t − 1), θ out (t − 2), . . . although not yet totally at the full scale; the subproject is
• the solar irradiation I(t) at hour t and the solar called the ‘Virtual Housing Laboratory’ (VHL).9 This is a
irradiation I(t − 1), I(t − 1), . . . system for simulating total energy use in the single-family
housing stock. The system is based on a sample with very
The following ARX model (AutoRegressive model with detailed data for 737 single-family houses and actual climate
eXogeneous inputs) with 11 free parameters has been data from nearby weather stations. The VHL is part of the
used: Energy Barometer project, which in its full version implies
collecting energy and climate data on an hourly basis for
w(t) + a 1w(t − 1) + a 2w(t − 2)
about 1,000 sampled houses.
= c 1 + b 1 θ in (t) + b 2 θ in (t − 1) + b 3 θ in (t − 2)
+ b 4 θ out (t) + b 5 θ out (t − 1) + b 6 θ out (t − 2) THE TECHNICAL SOLUTION USED TO COLLECT THE DATA

+ f 1I(t) + f 2I(t − 1) + f 3I(t − 2) + ε(t) The specification of the type of sensors used and the accuracy
of measurement are presented in Table 1.5. Reference [10]
Reference [7] also describes the way to generalize the
describes the specifications of the sensors used for a similar
results obtained for the houses investigated, the selection of
type of investigation.
the houses, the estimation of the total energy use for heating
The description of the measurement protocol and of
and of the annual change, the calculation of the error in the
the material used to collect the data is summarized in
estimates and the sample size.
reference [11]. This article presents the concept of the
The Energy Barometer system gives a solution to the
Energy Barometer used in Sweden with an emphasis on
combined problem of obtaining (i) timely and (ii) reliable
the data handling system and the associated communication
estimates of end-use consumption.
infrastructure. We summarize here the main characteristics
of the system.
• Timely estimates. Statistics about building energy end-use
The data handling system is based on a communication
have hitherto often been based on postal questionnaires
device, a residential gateway (‘E-box’) constructed by
with obvious problems of measurement quality and have
Ericsson Radio Systems. The practical reasons for using this
been reported more than one year after the energy use
system were:
occurred. With the method presented, using Internet
and modem communication systems, data can be
collected continuously and the quality of the • Intrusion into the home environment should cause
measurements can be specified and controlled. In minimal inconvenience for the residents in terms of,
addition, climatic corrections can be made to the for example, wiring and extensive installation work.
estimated annual energy use. • The system should be versatile and open enough to
• Reliable estimates. The reliability of the estimates allow other uses of the installed communication
depends on statistical procedures and the assumption facilities and hence decrease the part of the cost related
that the selected houses represent the population of to the measurements necessary for this project.
single-family houses. The results obtained indicate that • The system allows remote operation and maintenance to
a time step of 24 hours is too short for the dynamic avoid travel costs etc.
properties of a house to be averaged out and/or that
the houses do not behave according to a model with a The E-box is basically a computer, the operating system of
time-constant parameter system. This means either that which is specifically designed to allow a number of different
static models should be based on weekly data or that applications to run in parallel. Practical information and
models with time-varying parameter systems should be the specifications of the E-box system can be found on the
tested. The recommended length for the observation Ericsson’s website.12
period is about 7 to 10 weeks. In each house the sensors described above are installed.
They are connected via a network to the E-box. LonWorks
Reference [8] describes the way in which houses equipped has been used as local network since it enables physical-layer
with heat pumps can be taken into account in the Energy communication on the 220 volt cables already present in
Barometer system. the houses. The sensors are then simply plugged into the

Table 1.5 Details of the sensors used for measuring different variables and their accuracy
Variable Unit Accuracy Sensors and additional units
Energy
Electricity (electric heating, water-borne heating, heat pump) kWh ±2% kWh meter with output of pulses
District heating kWh ±2% Integration meter with output of pulses
Oil burner Litre ±5% Flow meter with output of pulses
Run time: oil h ±1%/h Sensor or electromagnetic sound
Temperature: indoor, outdoor ◦C ±0.3◦ C Temperature sensor
8 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Sampled g.A.Isol Qheat

house
#i LonWorks
Weather
data A gateway
‘E-box’
Text Hv U.A Text

Internet

Cint
Internet building service provider System for analysis and presentation

– The Energy Barometer – Pre-processed data

– Other building-related services

Figure 1.3 The structure of the Energy Barometer system Figure 1.4 A simplified single-zone model of a dwelling

wall and no extra wiring is necessary. Figure 1.3 shows the Since the use of light measurement techniques and methods
structure of the EB system. Raw energy and temperature is required for large-scale building certification, it has
data are collected continuously and stored in each local been chosen not to apply co-heating, nor to control the
E-box and transferred to a central server on a diurnal basis surrounding ambiences such as the climate in the basement
via the Internet. Weather data are sent from the nearest or the attic.
official climate station. The heated space is treated as a single zone (internal doors
From each house the following data have to be collected: are assumed to be open).

• the hourly energy use for heating (space and tap water)
DETERMINATION OF A TEST PROCEDURE
• the hourly energy use of energy for household appliances
• the hourly average indoor temperature. A constant indoor temperature strategy is applied. The
advantages are:
In addition to these data, technical data on, for example,
occupancy, heated area, year of construction, are also • It is easier to obtain the homogeneity of indoor
needed. This type of data is collected when the house is conditions.
inspected in connection with installation of sensors and the • It is applicable to all types of dwellings where energy
E-box. Reference [13] gives a sample inspection protocol. consumption is easily measurable (gas- and
electricity-heated houses).

THE SAVE HOUSE ENERGY LABELLING The disadvantages are:

PROCEDURE (HELP)
This procedure has been developed in the framework of the • Determination of the internal capacity is almost
EU-financed Save HELP project. impossible.
• The accuracy of the heating power measurement is
reduced; there is no continuous measurement of the
Theoretical background
consumption.
Reference [14] describes the theoretical method used.
The first objective was to characterize the energy
THE CHOICE OF MEASURED SIGNALS
performance of non-occupied buildings and the following
procedure was developed: The sensors and measurements were as follows:

1. Light monitoring of the studied dwelling according to a • a few temperature sensors in the heated zone
standardized test procedure. • one measurement of the outdoor temperature
2. The collected data set is then used to identify a • one extra temperature sensor in each unheated zone
simplified single-zone thermal model (as presented in (basement, attic...); not used for analysis but for
Figure 1.4, a global UA value, gA value and possibly C int , checking the test conditions
the internal capacitance). • measurement of the air change rate with
3. Once the model is obtained, a simulation can be carried PerFluorTracergas (PFT)
out with defined climatic and internal-gains data sets in • measurement of solar radiation with photovoltaic (PV)
order to determine a standardized energy consumption elements
level, which can be used as an energy performance index • periodic measurement of the energy consumption
for the considered building. (if possible, on a daily basis).
 Review of selected theoretical and experimental techniques for energy characterization of buildings 9

THE METHOD OF ANALYSIS next step in the development of the method has been to
set up an experimental environment in an occupied house
The evaluation of the method has shown the significance of
without disturbing the regular life of the occupants.
a continuous measurement of the temperatures and solar
The main aim was to develop a reliable monitoring
radiation, while, for practical reasons, energy measurement
procedure, which provided the necessary input for an
is performed on a daily basis and only an average air change
analysis tool. This analysis tool provides the information
rate for the heated space is obtainable.
that makes it possible to arrive at a normalized annual
As the analysis method is intended for use on a large scale
heating consumption for the dwelling, based on restricted
and because there is no need to use complex analysis tools
data from the heating season.
for identifying a pseudo-steady state model, a simple multiple
In the applied model, the overall UA value of the dwelling
linear regression model of the daily building energy balances
and the gA value are determined by means of multiple
has been adopted:
regression. For a good determination of these values, daily
 
measurements of the ambient temperature, the mean inside
Eday = UA · Ti,e − gA · Isol ,
temperature, the amount of heating inside the dwelling, the
day day
solar radiation and the mean ventilation rate must be avail-
with able. A multiple regression calculation is used to determine
 the unknown parameters (UAglobal and gA) from the model:
Eday = Qheat − C · Tday,day −1 − n · V · ρair · Cpair · Ti,e  
day Eday = UA global · Ti,e − gA · I sol
  day day
where day Ti,e and day Isol are the daily integrated indoor–
outdoor temperature differences and the daily integrated
global solar radiation measurements; the Eday variable is with
the difference between the daily energy consumption, the 
possible daily variation of the energy stored by the internal Eday = Qheat − C · Tday,day −1 − n · Vρ · Cp ·  T i,e
thermal mass, and the daily ventilation losses (based on a day

test duration averaged air change rate n). Once the data
set has been collected, each measurement day provides a This model can only be used if it is possible to define a
 
value for Eday , day Ti,e and day Isol , and the multiple linear unique Tint . This is possible only when the dwelling is
regression may be calculated, giving values for UA and gA, assumed to be one heated zone. In this case Tint is simply
the equivalent global UA value and the pseudo-solar gain the only temperature of the heated zone.
factor. Beyond its intrinsic simplicity, this method has the The experimental method applied does not require
advantage of allowing a direct and reliable evaluation of the detailed information on the building, but a relatively
confidence intervals of the regression parameters and gives long monitoring period (at least two weeks) is required
the possibility of calculating the confidence interval of Eday and energy readings can cause problems to, as well as
 
for any arbitrary chosen values of day Ti,e and day Isol . disturbances resulting from, the occupancy.
The evaluation of the results of experiments carried In spite of its uncertainty level, the experimental method can be seen
out in Belgium (Belgian climate) shows that, although the as the most reliable way to determine a normalized annual heating
total uncertainty level of the simplified method may be energy consumption since its output is relatively independent of the
considered to be a bit too high, mainly because of pessimistic observed conditions and since a very limited number of assumptions
assumptions, the following conclusions may be drawn in the have to be made regarding the building. It is, however, clear that
case of an well-insulated dwelling: such a method could hardly be applied on a large scale in the
framework of energy certification because of the amount of work that is
• The test duration should be about three weeks. required.
• The test should preferably be undertaken during the
winter, between the months of November and February,
and should not be undertaken in either October or Experimental procedure
March, unless the evaluation of the impact of a slightly The experimental approach adopted in the Save HELP
higher indoor constant temperature (e.g. 25◦ C) would project is a measurement-based approach (identification
lead to an acceptable uncertainty level. method).16 The approach relies on the real behaviour of
• The fact that the experimental confidence intervals on the building, which can be very different from that of its
the simplified consumption are rather larger than the design. It also allows the influence of the climate to be
statistical confidence intervals indicates that there is distinguished from that of the occupants.
nothing to be gained from trying to develop a more The method adopted is a time-integrated method derived
complex dwelling model whereas more efforts should be from a static heat balance of the building referred to as a
invested in increasing the accuracy of the measurements. non-controlled method because it can be applied without
any constraints on the heating schedule.
Reference [15] gives a global summary of the method Theoretical studies have been carried out to determine
developed and of the experimental techniques applied. The which parameters should be monitored. It appears that
10 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 1.6 The field experiment protocol

Variable Recording Time step Destination Caution
Energy Reading by the occupants Daily Heating and appliances Same hour each day
Indoor temperature Wireless (e.g. Tinytag) ≤1 hour Representative rooms Avoid heat sources and
recording sensors solar radiation
Air change rate Perfluorocarbon tracer (PFT) Integrated Each room Specific protocol defined
measurement by the analyst
Solar radiation Pyranometer ≤1 hour Horizontal Avoid shading
Outdoor temperature Recording sensor (e.g. Tinytag) ≤1 hour Outside close to the building Protect against radiation

measuring the magnitude of the indoor air temperature, (a)

the air change rate and the power heating should lead
to less than 10% uncertainty on the heating consumption,
the residual error being attributed to the random behaviour
of the occupant(s).
For global measurements in occupied houses the required
instrumentation should provide the following data:

• temperature measurements
• ventilation measurements
• climate data measurements or data from a local weather
station
• energy consumption data from measurements and/or
readings.

Temperature measurements were carried out with small

self-maintaining devices from Orion (Tinytag), which were
able to record data at 10 minute intervals for more than
a month. Five sensors were placed in the house and
one device was installed outside the house to measure
local ambient temperature. The measurement error in the
temperature is 0.2◦ C.
Tracer gas measurements using PFTs were carried out
in several occupied houses. The required equipment came (b)
by ordinary mail from PENTIAQ (Sweden). After the
measurement period the sampling tubes and other sensors
were sent back to PENTIAQ for analysis.
A description of the field experiment protocol is given in
Table 1.6 and pictures of the wireless recording sensor and
the PFT sources and samplers are shown in Figure 1.5.
General observations that can be made from the
experiment realized:

• Occupied houses give larger uncertainties than

unoccupied ones.
• Large errors were obtained for houses where large solar
gains can be assumed.
• The method cannot be applied to houses heated with
fuel (unless special techniques are used to measure
the fuel consumption).

Figure 1.5 (a) Wireless recording sensor; (b) PFT sources and samplers
STEM & PSTAR∗
This method has been developed in the USA. It building. The requirements concerning the experimental
consists of an intensive three days’ monitoring of the technique applied are very strict. References [17–19]
give a presentation of the method. The experimental
* ShortTerm Energy Monitoring (STEM) and the Primary-
techniques are described and the theoretical background
and Secondary-Term Analysis and Renormalization calculation and calculation methods are presented. We summarize here
procedure (PSTAR) the main characteristics.
 Review of selected theoretical and experimental techniques for energy characterization of buildings 11

Theoretical background
A wide spectrum of methods can be used in thermal
monitoring of buildings. The most useful methods can be
broadly classified as macrostatic and macrodynamic.
Macrostatic methods are based on time integration
of the energy balance of the building, with the input
data (building performance and driving functions such
as outdoor temperature) being similarly integrated. These
methods are simple and have been used for many years in a
wide variety of applications. However, these methods require
long-term data, and low information content can make it
Figure 1.6 Typical three-day STEM test protocol
difficult or impossible to reach reliable specific conclusions.
Furthermore, occupancy or other schedule changes during
the data period can be a major problem. the furnace is off and all heat comes from several portable
Macrodynamic methods directly employ the dynamic electric heaters individually switched on and off by the data-
energy balance equation of the building, with system acquisition computer. After midnight on the third night, heat
identification techniques used to extract some subset of is supplied from the installed heating system, operating in
the energy-balance parameters. These methods require response to the normal house thermostat.
some instrumentation in the building, such as an energy For analysis, the ‘renormalization factors’ are introduced
management system or a dedicated data logger. However, as simple multiplicative factors for the primary energy
the requisite testing data can be acquired in a few days. flows. Reference [20] describes the theory relating to
The effective leakage area is determined once at the the renormalization procedure. This procedure consists of
beginning of the procedure. A data acquisition system defining series of ‘time windows’, extracting a subset of
is temporarily installed in the building. Each channel is the parameters in each window sequentially with simple
scanned every 10 minutes and the hourly averages are linear least-squares regression, and iterating to convergence.
stored and transferred via modem for analysis. Typical data Primary- and Secondary-Term Analysis and Renormalization
channels are: (PSTAR) provides a mathematical formalism for separating
building energy flows, making it possible to identify the
• six for the inside air temperature three primary thermal characteristics:
• two for the outside temperature
• one or more for the buffer space temperature • the building loss coefficient
• once each for the outside relative humidity, global • the effective building mass
horizontal solar radiation, global vertical solar radiation, • the effective solar gain area.
total electrical power, wind velocity, furnace switch-on
time and surface heat flux in case of contact with the It then uses the adjusted model to predict future building
ground. performance. In the PSTAR procedure, the heat flow into
the room air is mathematically separated into several terms
The test protocol is quite simple. Each of the primary heat according to the effect causing the heat flow. The terms are
flows is forced to become dominant in the energy balance as follows:
for some time period in the test. First, the steady-state
conduction term is forced to be dominant by inclusion • Primary terms (to be renormalized):
of one or more nights when the interior temperature is Q 1 building loss coefficient (BLC) times outside–inside
steady (no night setback). Second, the internal mass storage temperature difference. This is the building
term is forced to be large by inclusion of a night of conduction gain to the room air from the outside
temperature decay (setback). Finally, during the test, one air under steady-state conditions. BLC is determined
or more relatively sunny days must occur to ensure a good from the model.
renormalization of the solar gain term. Q 2 heat flow to the room air due to a change in the
The test procedure is programmed into the data- inside air temperature. Calculated from the model.
acquisition computer and requires four days and three Q 3 heat flow to the room air due to solar gain. This
nights, including set-up, take-down and analysis (Figure 1.6). includes the effect of solar gains through windows,
The objective is to obtain steady-state conditions during heat stored in the internal mass of the building
the first night, to do a cool-down test on the second night that is subsequently discharged into the room,
and to calibrate the heating system on the third night. and heat flow through the external walls due to
The tests during the second and third nights are started at solar absorption. Q3 is calculated using the
midnight after a steady-state lead-in period. Daytime data simulation model by setting the inside and outside
are used to determine the effect of the solar gains. As far temperatures equal and constant; the calculated
as practical, all house appliances and lights are turned off cooling load is then the heat flow due only to
during the entire test. Up to midnight on the third night, solar gain.
12 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

• Primary terms (not renormalized): This method has been successfully applied over a wide
Q4 measured heat flow to the room air due to internal range of climatic conditions. A great potential exists for use
gains. This is all the electrical energy into the of the method for evaluating other thermal effects, such as
building, including from the electric space heater. the effects of the occupant(s), fireplace efficiency and the
Q 5 heat flow to the room air due to heating of effectiveness of thermal storage strategies. This method can
infiltration air. Calculated using the Sherman– also be applied to test the efficiency of the cooling strategies
Grimsrud model based on the leakage area; the realized by ventilation.
inside–outside temperature is estimated based on A major drawback of this method is that the experimental
the measured outside temperature and the relative conditions are very strict and require that the building is unoccupied.
humidity. The method also needs a short but intensive effort involving
• Secondary terms (not renormalized): personal time.
Q 6 heat flow to the room air due to a change in the A study describing the accuracy and the repeatability of
outside temperature. Calculated from the model. the results obtained with this technique is presented in
Q 7 extra heat to the room air due to the lowering of reference [21]. The conclusions are that, compared with
the sky temperature below the outside air other methods, this technique makes it possible to obtain
temperature. quite accurate results.
Q 8 heat flow to the room air due to conduction from
an adjacent buffer space, such as a crawl space or Applications of the method
a basement.
Q 9 average heat flow to the ground due to direct References [22–24] present results of experiments realized
contact with the earth. on residential buildings.
A comparison can be made between the results obtained
Renormalization of the first three terms is done step-wise as with the tests and the interpretation of the results obtained
follows. The energy balances descibed are averages over the in four houses compared with the energy consumption
time period. In each step, the previously determined values found by calculation methods. It is shown that important
of renormalization constant are used. differences can be found between the two techniques and
that this method can make it possible to identify the building
1. During a period of 2 to 4 hours at the end of the night, characteristics more effectively. Some explanations are given
when the inside temperature is maintained reasonably to explain the difference between the purely calculated
steady (called the co-heat period), Q 1 , Q4 and Q 5 are the results and the energy consumption obtained with measured
dominant terms. That is to say the heat input from the data. The influence of the solar radiation is always important
electric heaters should approximately balance the heat in the interpretation of the results.
losses by conduction and infiltration because heat The STEM and PSTAR techniques can also be applied
storage, solar and other effects are small at this time. with some adaptation on commercial buildings. References
The building loss coefficient renormalization factor [25–26] present the results of two experiences realized on
p1 is determined as necessary to reconcile the this type of building. In the case of the commercial building,
observations. Specifically, Q 1 is multiplied by a one of the major weaknesses of this implementation of the
renormalization constant p1 to achieve an exact PSTAR approach is the single-zone nature of the modelling.
energy balance. This point is not developed here since the objective of this
2. During the cool-down period, the primary heat flow study is oriented to residential buildings.
into the room air results from the discharge of the The method has also been applied to estimating the effi-
building mass, because electric heat input, Q4 , is zero, ciency of passive cooling strategies on residential buildings;
or at least quite small. The mass renormalization factor the results of a case study are given in reference [27].
p2 is determined as required to reconcile the heat
balance over the entire analysis period. Specifically, an NEURAL NETWORKS
energy balance is achieved by multiplying Q 2 by a
Background theory
renormalization constant p2 .
3. During the daytime hours, a major heat input is from The application of neural network techniques is very wide.
solar gains, and electrical heat is correspondingly In the scope of this study, neural network techniques are
reduced. The solar gain renormalization factor is used to predict the future thermal behaviour of a building
determined, as required, to reconcile the heat when sufficient data from the past are known. References
balance over the entire analysis period. Specifically, [28–30] give short but comprehensive descriptions of the
an energy balance is achieved by multiplying Q 3 by theory relating to the neural networks, we summarize here
a renormalization constant p3 . the main characteristics of the technique.
Conceptually, a neural network (NN) is made up of
Steps 1 to 3 are repeated until the renormalization constants interconnected nodes arranged in at least three layers
p1 , p2 and p3 stabilize. The sum of Q 1 through Q 9 (with the (Figure 1.7).
first three terms renormalized) is the energy imbalance Qnet . The input layers merely receive the data patterns related
This should be small throughout the test period. to the input variables. The number of input nodes
 Review of selected theoretical and experimental techniques for energy characterization of buildings 13

References [28] and [32–33] present the results of

case studies. The method applied uses NN techniques to
obtain models that are capable of predicting the supplied
heating demand of small family buildings. The models
are based on a limited amount of measured diurnal
performance data for a short time period. The results
obtained show that, with access to only measured data of
supplied space heating demand and climatic data in terms
of indoor and outdoor temperatures, the supplied space
heating demand can be predicted to within 5–10% on an
annual basis.
The method has been applied on 10 single-family houses
Layer
Input La Hidden layers Output layer
situated 700 km north of Stockholm, Sweden. Limited
Figure 1.7 Neural network scheme
monitoring has been realized. The data coming from
the monitoring were meteorological data and building
performance data; energy demand, outdoor and indoor
consequently equals the number of measured data values. temperatures were measured at the buildings. Other mete-
The hidden and output layers both actively process data. orological data, such as wind, relative humidity, air pressure
To do that, each neurone acts as a connection. Associated and solar irradiation, were measured at a nearby airport.
with each connection is an adjustable value called a weight. This method is a combination of a neural network and
Basically, a node calculates the weighted sum of its input, a quasi-physical description, which requires only access to
then passes the sum through a function to produce a the average daily outdoor and indoor temperatures and to
result. This transfer function is typically a steadily increasing the space heating demand for a limited period of time.
S-shaped curve. The attenuation at the upper and the lower Despite the diversity of the different buildings in term of
limits of the S constrains the sum within fixed limits. The building design, ventilation system, control system and local
back-propagation algorithm is able to change the values of climate, this approach yielded a rather robust performance.
its weight in response to errors. It then compares the desired Except for periods when the supplied space heating demand
and the actual results. was very small, the approach yielded fair predictions. The
The differences are the errors in the output layers, which method requires only performance data on a level that can be
the network passes back to the hidden layer using the same supplied by most house owners. The method can also be used
weighted corrections. After each output node and each for both the domestic and the total energy demand of
hidden node finds its error value it adjusts the weight to a building.
reduce its error. The equation that changes the weights – The traditional method for simulating the energy demand
called the delta rule – is designed to minimize the network’s in buildings is based on implementations of building
sum-squared error. After training, the network should be characteristics in more or less ideal physical sub-processes.
tested with known data that has not been used during Although this method gives a good physical understanding,
training. The accuracy of the network with a pattern outside it requires a large number of parameters and different
the training set is called generalization and indicates its degrees of idealization and simplification. Other common
reliability in an application. After training and testing, the modelling techniques based on measured performance
network is ready to process unknown data. Basically, applying data are, for instance, time-series analysis and statistical
a pattern to the input produces a corresponding pattern methods. These methods require no detailed knowledge
at the output. The network therefore acts as a model of of the building, but, instead, the results can be difficult to
a function, mapping input patterns to output patterns. It interpret in physical terms.
learns this association solely from the training data, even if An advantage of neural network techniques is their good
the equation describing the function is non-linear, unknown ability to map non-linear dependencies between input and
or both. output data and, also, that this can be done without any
In the summer 1993, an energy modelling competition conceptions of intrinsic relations in the presented data.
was organized by ASHRAE, called ‘The great energy A drawback of neural networks, as well as many other
predictor shoot-out’.31 The purpose was to evaluate different non-analytic models, is the difficulty of interpreting the
methods for predicting hourly energy use. The competition model in practical terms, as well as the limitations in
was supplied with four months of historical data and was the accuracy of the predictions for events outside the
asked to predict the following two months. Five of the six data sets.
most successful contenders used different neural network Significant for neural networks is that they do not
techniques. demand any explicit rules or knowledge for processing data.
For neural networks, the rules are built within the system.
With this technique the supplied space heating demand
Applications of the neural network technique can be predicted with an accuracy of about 5%. Although this
Neural networks are used in some cases to determine technique does not need heavy monitoring and the results have a
experimentally the annual space heating demand. fair accuracy, the exploitation of the data requires experience and
14 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

feeling and is relatively time-intensive. The model needs first to OTHER RELATED TOPICS
be trained with the available data before it is used with the
meteorological conditions of the reference year. This method In this section we summarize some other interesting articles.
could be difficult to apply for certification purposes.
Reference [34] presents a similar approach. In this Miscellaneous
model, the behaviour of the inhabitants in terms of the
domestic load is included. In order to do this, the predicted Reference [39] describes the results of a set of simulations
annual variation of the domestic load based on measured of the total energy consumption of two buildings by taking
performance data, together with an assumption of an into account the uncertainties of the different parameters
annual profile of the domestic load obtained from a more playing a role in the energy consumption (climate, building
extensive measurement, is used. In addition to predicting and inhabitants). These simulations and the statistical
the supplied space heating demand, the method predicts utilization of them make it possible to assess the influence
the total energy demand. As model input, the model uses of these parameters on the total variability of the energy
the difference between indoor and outdoor temperatures, a consumption. Some interesting results are that, without
building climate perspective obtained from dynamic energy knowing the inhabitant behaviour, it is impossible to predict
simulation software and a measure of inhabitant influence in the total energy consumption more accurately than ±15–
terms of a predicted domestic load. The predictions of the 20%. Furthermore, the heating and ventilation energy has
domestic load were based on measurements performed in an uncertainty of ±25–40% if the inhabitant’s behaviour is
a separate single-family building. Compared to a previous unknown. The colder the climate, the better the accuracy of
model, based mainly on a building-climate perspective, the heating consumption.
the deviation between the predicted and measured annual This article shows that certification of the energy
variation decreased from 7.5% to 4%. consumption by a descriptive method will unavoidably be
Neural networks can also be used to analyse the affected by the uncertainties in the different parameters
interactions between the heating system and the energy (climate, building and inhabitants).
load35 and to analyse the interactions between heating and Reference [40] describes the origin of the error
domestic load in an occupied single-family building.36 encountered when energy simulations are realized with
energy signature models. It appears that including the solar
radiation in the model can significantly improve the accuracy
Supplementary examples of applications of of the results obtained.
the neural network technique The article shows that the calculation and the introduction
Neural networks can also be used in other types of of the weather correlation of solar radiation I and outside
application. temperature T in the simplest model for every frequency
Reference [29] presents a case study of an application of of data confirms quantitatively that omission of the solar
the neural network technique to predicting the building radiation is the main cause of error in the estimate of the
energy consumption of a building in the future without parameters of the building.
knowledge of the immediately past energy consumption. Conventional building load simulation models give
Such a prediction is of value when estimates of what performance predictions based upon physical properties of
a building, retrofitted with energy conservation features, materials and specifications of equipment, the real values
would have consumed had it not been retrofitted. of which are frequently uncertain. Interactions between
Reference [37] presents a supplementary application the building and the inhabitants or the environment
of neural networks. It is used here to produce a are also imperfectly known. In another approach, some
utility consumption prediction indicator for a commercial models use measured data; these models, sometimes called
building. The predictors are afterwards used in expert measurement-based energy models, extrapolate building
systems to realize diagnostics for Heating, Ventilation and parameter estimations and energy performance predictions
Air Conditioning (HVAC) systems. The procedure used to over an extended period. The available data may only consist
find the predictors uses historical data, such as weather data, of energy consumption and weather characteristics.
building occupancy and other factors known to affect energy
consumption.
Experimental techniques
Reference [30] presents an application of neural networks
in the case of the control of the heat power delivered to a There are some other references to experiments realized
building. Four parameters have to be introduced as input using monitoring techniques (see references [41–47]). The
to the NN: outdoor air temperature, solar radiation, indoor purposes of these tests can be very different and some articles
air temperature and energy consumption at time t − 1. The are interesting but less relevant to this study. We do not
result of the NN is the energy consumption at time t, which is summarize these articles here.
the heat power. The NN technique delivers accurate results
and could be developed for other control purposes.
The last application of neural networks presented here
Calculation methods and parameter estimation
concerns the use of the CO2 concentration to predict energy We give here some other references to calculation methods
consumption.38 and techniques for estimating the parameters used in the
 Review of selected theoretical and experimental techniques for energy characterization of buildings 15

calculation methods. Since this is not the main topic of this rate. The technique used (PFT) requires care, needing
report, we do not give much detail. qualified personnel with experience and a sensitive
Reference [48] describes an experiment carried out to test approach.
and validate a measurement-based model with ‘synthetic’ The procedure developed in the STEM and PSTAR
consumption data generated by a conventional building method is very detailed and much relevant information can
load simulation model. The model tested is SUBMET. This be found in the literature. The protocol used is very strict
technique makes it possible to test the algorithm, to define and the houses have to be unoccupied in order to make
what is the best form to give to the data to obtain the best measurement convenient. The results obtained are probably
accuracy, to define the best accuracy that can be expected the most accurate of the different methods described. This
from the model, etc. Note that this technique could also type of prescriptive measurement protocol can probably
be used to test measurement-based models for cooling not be adopted in the developed methodology because
energy use. of the complexity of the set-up and of the calculation
The article begins by giving a description of different phase.
existing measurement-based building energy models. The experimental techniques used in the neural network
Measurement-based models typically derive parameters methods have essentially been developed in order to
of energy performance from the measured data. These provide data for the calculation methods. However, it should
parameters are combined with weather data and other be noted that the type of networks adopted can influence
physical characteristics of the building to predict thermal the results. With these types of technique, the choice of
performance or average annual space heating consumption. the data used to train the network can also have an influence
There are both dynamic and static measurement-based on the results obtained. For each case it is necessary to
models. Among the dynamic models is BEVA (Building adapt/train the model used for that specific case. It should
Energy Vector Analysis). Static models PRISM and SUBMET also be noted that the results obtained can be difficult
aim to predict the average space heating use in new, to interpret in a physical sense. These types of technique
low-energy single-family homes. SUBMET is a regression- require highly qualified personnel and are probably still
based model but requires more detailed information than best kept for research purposes rather than for certification
PRISM. methods.
Reference [49] presents a comparison of different Some other methods and related topics have been
methods for dynamic analysis of measured energy use. The identified and are included in the references.
theory and the interpretation of the different methods are We have not been able to find articles describing methods
presented. The methods are thermal networks, the ARMA used to characterize the functioning of active cooling.
(AutoRegressive Moving Average) model, differential Some of the calculation techniques described could be
equations and modal analysis. extended for this purpose but no experiment tests have been
The same example has been tested with the four described.
models. The difficulties related to the four models are
presented. This article gives many bibliographic references
to publications describing each method separately. The
article deals more with calculation methods than with the Part 2. In situ evaluation of UA
experimental techniques used to collect the data.
Further information can be found in references [50–59].
and gA values – an overview of
possibilities and difficulties
INTRODUCTION
CONCLUSIONS This part of the report gives an overview of the state of the
There are several experimental methods that could be useful art in relation to the on-site evaluation of the UA and gA
to the procedure currently developed. However, most of values of buildings.
the references that we have found were more focused on
the calculation methods rather than on the detail of the
experimental monitoring used.
DIRECT MEASUREMENTS OF U-VALUE
Because of the various advantages offered by the It is possible to measure with quite good accuracy (5–10%
methodology used, the technology of the Energy Barometer or better) the U value of building components. Such
project seems to be the most appropriate for the method measurement is typically done by using a heat flow meter
currently developed. The use of the Internet and a modem installed on the inner surface of the component and
to collect the data is the most promising technology temperature sensors on both surfaces.
available for developing an accurate method with reasonable The procedure for determining the ‘surface-to-surface
exploitation costs. thermal resistance’ is described in the European standard
The measuring techniques employed in the Save prEN 12494.60
HELP project is a light monitoring approach that could It should be noted that, in the case of transparent
easily be employed in other circumstances. The only components (glazing, etc.), it is not easy to carry out
sensitive point here is the measurement of the ventilation a measurement if there is incident shortwave radiation,
16 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

because of problems of accuracy of the measurements As far as the authors are aware, the procedure was
realized (e.g. glazing surface temperature). probably unique at the time it was first applied in the
PLEIADE dwelling, when specific attention was given to the
minimization of the uncertainties. Basically, it consists of
Measurements in climate chambers four essential features:
HOT BOX APPARATUS
1. A refined procedure for monitoring climate parameters and
The hot box measurement technique using a guarded or energy consumption data. The monitoring measures not
calibrated hot box (EN ISO 8990/EN 1946-4)61 is well known only all room temperatures, the outdoor temperature,
for identifying the UA value of building components. the solar radiation and energy use, but also the air flow
rates with multiple tracer gas measurements and the
OUTDOOR TEST FACILITIES water vapour balance of the house.
Since the 1980s, outdoor climate chambers have received 2. Physical measures for minimizing solar gains during the
increased interest: monitoring. Based on the error analysis in simulated
experiments, evidence was provided that carrying out
experiments involving important solar gains (as is the
• In the USA, the MOWITT unit developed at Lawrence case in PLEIADE) in combination with low transmission
Berkeley Laboratory (LBL) has to be mentioned. losses would lead to a very large uncertainty in the
• In Europe, efforts have been concentrated in the predicted thermal transmission losses for the whole
framework of the European PASSYS, PASLINK building. This uncertainty can be significantly reduced
(Figure 1.8), COMPASS, PV-HYBRID-PAS and IQ-TEST by minimizing the solar gains. In practice, this was
projects. done by having self-adhesive white sheets on the
outside of the glazing and by putting up a scaffolding,
covered by an opaque canvas, around the façades
Detailed monitoring campaigns in (see Figure 1.9).
non-occupied buildings 3. Refined control of the indoor climate and of certain boundary
conditions. The approach used in NBN B62-301 for
GENERAL
determining the transmission losses at building level
The verification of thermal performances in real buildings gives different weighting factors for the losses through
seems very attractive. In the framework of the PLEIADE the components in direct contact with outside
project,62 one of the aims was to verify in situ the reliability (weighting factor WF = 1), through the floors in
of the calculation procedures for estimating the global contact with the basement (WF = 2/3) and through
transmission losses at building level, calculated according to the common walls (WF = 0). A simple co-heating
the Belgian standards NBN B62-301, NBN B62-002 and NBN experiment does not allow a distinction to be made
B62-003.63,64,65 between the three types of heating losses measured

Figure 1.8 A picture of a PASLINK test cell

 Review of selected theoretical and experimental techniques for energy characterization of buildings 17

Figure 1.9 Specific measures in the PLEIADE building for increasing monitoring accuracy. Source: BBRI66

45
Text
40
35
30
25
20
[˚C]

15

Tcommon walls 10
Tin 5
0
0 5 10 15 20 25 30
-5

Qheat -10 [Days]

Tbasement Tindoor Tbasement Tcommon walls Text

Figure 1.10 The model assumed for the building (left) and the heating regime (right). Source: BBRI66

during the experiment. In order to allow a correct

weighting, a methodology was developed to have some
control of the temperature profiles in the basement
(Tbasement ) and of the outside surface temperature of the Transmission losses
5,700 MJ
common wall (T common walls ) of the PLEIADE dwelling. 71%
The latter was achieved by integrating electrical heating
foils into the construction of the PLEIADE dwelling
(Figure 1.9). As is illustrated in Figure 1.10, very
different temperature profiles were used so that there
Drying Ventilation losses
was a low correlation factor between the various power 1,600 MJ
temperature profiles. Moreover, high internal 700 MJ 20%
temperatures were applied to increase the temperature 9%
difference between inside and outside.
4. The use of system identification techniques for data analysis. Figure 1.11 Various energy flows during the monitoring period in the
Based on the methodologies developed in the PASLINK PLEIADE dwelling. Source: BBRI66
projects, system identification techniques are applied to
the measured data. energy use served to compensate for the transmission losses.
The derived thermal insulation level (according to NBN
PRACTICAL RESULTS B62-30163 ) was K27* (95% confidence interval: ±4), which
agrees very well with the theoretical derived value of K28
For the monitoring period ( January 1995), the relative
(Figure 1.12). If the monitoring had been limited to the
energy use to compensate for the transmission losses, the
ventilation losses and the drying out of the dwelling (the * The K-level represents the thermal insulation level of a building.
monitoring period was less than one year after construction) For buildings with a low capacity, it represents 100 times the average
is given in Figure 1.11. Only 71% of the total identified U value of the building K = 100 × Um for buildings with V/A  1 M
 18 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

transmission losses, and no ventilation losses or energy use According to NBN B62-301, all walls on the border line
for drying out of the building assumed (Figure 1.11), a with the neighbours have to be considered as a common wall.
thermal insulation level K38 would have been calculated, In reality (and allowed by the urban regulations), one of the
a value 40% higher than the calculated one. neighbours has used another building form (Figure 1.13),
The probable conclusion would then have been that as a result of which a part of the assumed common wall
the theoretical calculations strongly overestimated the real becomes an external wall (about 25 m2 ). It is clear that such
performance. a modification can significantly increase the transmission
It is important to draw attention to the treatment of the losses (especially if this wall is not insulated).
‘common walls’ of the PLEIADE dwelling. In principle, the The authors believes that the experiences in the PLEIADE
thermal insulation level neglects the heat losses through dwelling as well as in other buildings and in the PASLINK
the common walls. The PLEIADE dwelling was built test environment make it possible to draw the following
before the adjacent dwellings were constructed. conclusions with respect to empirical model validation of
thermal simulation programmes:

35
K-level according NBN B62-301

• Rather good agreement between prediction and

30 K27 monitoring results can probably be found in many cases
K28
25 if a number of conditions are met: the overall
20 transmission losses have to be calculated based on the
real situation and take into account the precise location
15
of adjacent buildings; the insulation has to be correctly
10 installed and the assumed λ values should be based on
5 the measured values for the materials applied in the
building.
0
Predicted performance Performance level • Important differences may occur if certain practical
derived from monitoring aspects of the monitoring are neglected, e.g. the building
Figure 1.12 PLEIADE: Comparison between predicted and measured form of neighbouring buildings (see Figure 1.13),
thermal insulation levels. Source: BBRI66 thermal bridges, etc.

Figure 1.13 A ‘common’ wall in PLEIADE (Architect: P. Jaspad)

 Review of selected theoretical and experimental techniques for energy characterization of buildings 19

In the literature, (rather) important differences between solar gains

theoretical calculations and monitoring results are some- transmission losses
times reported. This would have also been the case for
PLEIADE if the ventilation losses, the drying out of the
building, the losses through the common walls, etc. had not ventilation
been correctly monitored. The authors suggest that one losses
must be extremely careful with statements concluding that
monitoring results have ‘proven’ that certain simulation
procedures are unreliable. Discrepancies are probably quite
often more related to the quality of the monitoring activities Figure 1.14 Energy flows involved in the identification process
than to the quality of the simulation tool.

Methodology
Detailed monitoring in occupied dwellings
The essential features of the methodology are:
BRIEF OVERVIEW OF INTERESTING STUDIES
Save HELP
• detailed continuous measurement of indoor and
The Save HELP project has already been discussed earlier outdoor temperature
in this chapter.14 Here we focus more on the results of the • on-site measurement of solar radiation
calculation method. • estimation of the shading impact of the environment for
This European study describes a simplified method for the various windows
the energy certification of unoccupied buildings. First, • estimation of the ventilation rate (by the passive tracer
the applicability of the proposed method was assessed gas method, by the pressurization method in
by means of numerical simulations, making it possible combination with the LBL-model, etc.)
to define the experimental requirements. Second, the • daily recording of energy consumption
sensitivity study focused on the seasonal, duration and • estimation of the thermal capacity of the dwelling.
building type dependency of the method accuracy. Third,
the resulting method was applied to a well-known
These measurements are typically made for 20–40 days.
unoccupied test dwelling, making it possible to check in
Based on daily average values for all parameters, system
practice the relevance and validity of the method and of its
identification is carried out, so that RC models such as
design.
the one presented in Figure 1.15 are identified. The main
The basic idea of the method is to derive a simplified
result of a single identification run is an estimation of the
thermal model of the dwelling to be certified by means of
UA and gA values together with their confidence intervals.
a limited monitoring of the building (energy consumption,
As part of the whole concept, correct estimation of
indoor and outdoor climate) with uncontrolled indoor
the confidence intervals is very important. A Monte-Carlo
conditions (constant temperature strategy). This simplified
approach is therefore applied, whereby for each input
model is then used to calculate, on the basis of typical climatic
variable a probability function is estimated. A hundred
data and a reference occupancy effect, a standardized
different data sets are generated (Figure 1.16) so that there
heating energy consumption for the heating season. As the
are 100 sets of UA and gA values with their respective
objective of the Save HELP project is to design a certification
confidence intervals.
procedure and to assess its relevance in a straightforward way,
several simplifying assumptions were made. Among them,
an electrically heated dwelling with a heating system having
an efficiency of 100% was used as the base case. Although
the method proposed indicates how to deal with lower
efficiencies, its applicability remains limited to buildings
equipped with a heating system that allows a daily reading
of the heating energy consumption.

PhD thesis by Zoltan Somogyi

An interesting approach has been developed in the PhD
thesis of Z. Somogyi,67 which takes a similar approach to that
of the Save HELP project.
This approach takes into account (Figure 1.14):

• transmission and ventilation losses

• solar gains
• estimation of internal gains. Figure 1.15 Example of an RC system used for system identification
20 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

In practice, one observes: The high correlation clearly indicates that it is not necessary
to make accurate estimations of the individual values of
• large confidence intervals UA and gA. A so-called normalized energy consumption is
• a high correlation between the different sets of UA and therefore determined, which estimates the annual energy
gA values. consumption for a standard year and for fixed indoor
conditions (Figure 1.17).

INPUT
PARAMETERS Error analysis
p RANDOM In Figure 1.18, the results are shown for an unoccupied
SELECTON dwelling in the Belgian climate. The large differences in
μ NOT accuracy can clearly be seen.
In Figure 1.19, the impact of the average indoor
MODEL STORE YES temperature is shown. The higher the average indoor
RUN=100
SIMULATION OUTPUT temperature, the better the accuracy.
OUTPUT On site, the method was applied in one unoccupied
p dwelling (Figure 1.20) and in two occupied dwellings
The output { UA, gA } (Figure 1.21).
are N(μ,σ) The results and confidence intervals for the normalized
μ energy consumption are shown in Figure 1.22.
Figure 1.16 The concept of the Monte-Carlo approach for error analysis

30 days test duration

40.0
MULTI-ZONE DYNAMIC
Relative Conf. Int. on ECN [%]

COMPARISON THERMAL SIMULATION 35.0 Confidence interval on normalized

VALIDATION OF
energy consumption is 18% for the
THE RESULTS 30.0
measurement period statrted on October
PSEUCO-MEASURED 26, with a duration of 30 days and an
SIGNALS 25.0
average dwelling temperature of 15°C
20.0
IDENTIFIED DYNAMIC 15.0
STANDARD BUILDING MODEL STANDARD
CLIMATIC UNCERTAINTY OCCUPANCY 10.0
DATA DATA
5.0
NORMALISED ENERGY 0.0
CONSUMPTION (ECN)
ct

ct

ov

ec

eb

ar

pr

ay
Ja

Fe
O

-O

UNCERTAINTY
-M

-A
-N

-D

-M
-F
1-

9-

3-
26

19
28

25
20

15

14
Figure 1.17 Determination of the normalized energy consumption and Test start date
related confidence intervals Figure 1.19 The impact ofTset
average C indoor
= 15°C
=15 Tset =25temperature
= 25°C
C on accuracy
Tset = 20°C

17.00-19.00
15.00-17.00
13.00-15.00
19.00 11.00-13.00
Relative Confidence on

17.00 9.00-11.00
15.00 7.00-9.00
ECN [%]

13.00 5
10 5.00-7.00
0
11.00 15
0
9.00 20
25
0
7.00
30
5.00
0 35
May
9-May
Apr
19-Apr
Mar
Mar
30-Mar
10-Mar
eb
18-Feb
an
29-Jan

40
an
9-Jan
c
20-Dec
v
30-Nov
v
10-Nov
21-Oct

1-Oct

Figure 1.18 The accuracy of the identified normalized energy consumption for various measurement periods (the y-axis
shows the duration of the measurement in days)
 Review of selected theoretical and experimental techniques for energy characterization of buildings 21

Very well insulated

Very good air-tightness
Non-occupied
Figure 1.20 The unoccupied dwelling used for monitoring

Dwelling 1 Dwelling 2 However, it is almost certain that there are major

differences in the relative accuracy:
South
So
facade
fac • In the north of Europe, in winter time there are high
South facade transmission losses (and large temperature differences)
and very low solar gains. As a result, the difference
between losses and gains can be estimated with a good
accuracy.
W t ffacade
West d • In southern warm regions, in winter time the absolute
East facade values of the losses are relatively low (and temperature
Poorly insulated Well insulated differences are relatively small). Moreover, there are
Limited glazing area Huge glazing area quite often rather substantial solar gains. As a result, the
Very leaky Good air-tightness difference between losses and gains cannot be estimated
Occupied Occupied with the same relative accuracy as in the first case.
Figure 1.21 The two occupied dwellings used for monitoring • In moderate climates, the situation is probably
somewhere in between the previous two cases.

The following conclusions are important: The authors therefore believe that the best relative accuracy
for the normalized energy consumption can be obtained in
cold climates.
• The best results are found in the non-occupied dwelling
(IDEE). This is clearly due to the relatively small
uncertainty in the internal gains, the ventilation rates INDIVIDUAL VALUES FOR LOSSES AND GAINS
and the average building temperature. Northern cold climates
• The relative accuracy is better in the poorly insulated
dwelling (ECSII). This is also quite logical since the In line with the observations given above, good relative
losses are relatively large, which allows a more accurate accuracy is expected for the loss value (transmission +
estimation of the normalized energy consumption. ventilation) in northern cold climates if measurements are
carried out during the cold winter months. Indeed, the
losses are almost equal to the heating demand. However,
The impact of climatic region relatively poor accuracy is probably found for the gA value
NORMALIZED ENERGY CONSUMPTION because of the low value of the solar gains, both in absolute
terms and in comparison with the losses.
The authors are not aware of any systematic studies relating
to the impact of the climate (e.g. warm region in Southern
Southern warm climates
Europe, moderate region in Western Europe or cold region
in Northern Europe) on the relative accuracy of the The measured energy consumption is often the difference
normalized energy consumption. between two values (losses and gains) of rather similar
22 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Energy Normalisation results from different monitoring periods

40,000.0
Normalised Energy Consumption [kWh]

13.1%
35,000.0 14.1%

30,000.0

25,000.0

20,000.0
21.
2 6%
21.6% 1 2%
19.
19.2% 27.5%
15,000.0
33.6%
10,000.0
20.2%
2
5,000.0

00
ECSII 1 ECSII 2 ECSI 1 ECSI 2 ECSI 2&3 ECSI 3 IDEE 1
Monitored dwelling and monitoring period
Calculated with ELAB
©1997,Z. Som ogyi

Figure 1.22 Results for one unoccupied dwelling and two occupied dwellings

Figure 1.24 The ESTI sensor

Figure 1.23 Temperature datalogger

SOLAR RADIATION DATA

In the past, on-site measurement of solar radiation data was

magnitude. It is therefore not possible to estimate the loss quite expensive. Here also, there have been some interesting
and gain factors with a reasonable accuracy. developments. In particular, the ESTI sensor (Figure 1.24),
developed by JRC Ispra, in combination with the data logger
technology mentioned in the previous section makes on-site
measurements rather affordable.
Monitoring equipment
TEMPERATURE
AIR CHANGE RATE
Long-term temperature measurements were in the past The air change rate can be measured with active or passive
only possible with rather heavy and expensive data loggers tracer gas techniques. In occupied buildings, active tracer
in combination with temperature sensors. During the last gas measurements are not possible. In such cases, passive
decade, however, small data loggers have become available tracer gas measurements (Figure 1.5) can be an alternative.
and very affordable prices are now combined with a large On average, the accuracy is of the order of 20–30%.
storage capacity and quite good accuracy. One can now buy,
for about m100 or less, data loggers that allow the storage of
ENERGY USE
up to 100,000 measurement points. Two typical data loggers
are shown in Figures 1.5 and 1.23. The data can subsequently The measurement of energy use can be difficult in some
be read by a normal PC. cases. A summary of the different adopted techniques
 Review of selected theoretical and experimental techniques for energy characterization of buildings 23

available to measure of the energy consumed is given by 4. Westergren K-E, 2000, Estimation of energy need for heating in
Akander, Johannesson et al.68 single-family houses, R&D-report No. 3, R&D committee, Royal
Institute of Technology, Stockholm.
5. Norlen U, 1985, ‘Monitoring energy consumption in the
Electricity and gas Swedish building stock’, Proceedings of Conference on Optimisation
of Heating Consumption, Prague, Swedish Institute for Building
In principle, daily recording is no problem. In practice, Research, Gävle.
monitoring equipment has to be installed and security 6. Hammarsten S, 1987, ‘A critical appraisal of energy signature
aspects have to be taken into account. models’, in Applied Energy, 26(2), pp. 97–110.
7. Hammarsten S, 1984, ‘Estimation of energy balances for
houses’, Bulletin M84:18; Doctoral thesis, The National Swedish
Fuel, wood, etc. Institute for Building Research, Gävle.
8. Westergren K-E, 2000, Energy use for heating in houses with
It is often not possible to monitor these.
a heat pump; Working paper series No 47 Research and
Development Committee at University of Gävle in cooperation
with The Royal Institute of Technology, Stockholm.
CONCLUSIONS 9. Westergren K-E and Waller T, 1998, ‘Virtual Housing
Laboratory, A system for simulating the energy use for heating
1. Direct on-site measurement of U value is possible.
in single family houses’, Working Paper No. 1, R&D Committee,
A European standard (prEN 12494) is available and
Royal Institute of Technology and University of Gävle,
describes the required procedure.60
Sweden.
2. Reliable identification of the UA and gA values in 10. Boman C A, Jonsson B-M and Mansson L-G, 1993, Eleffektiva
unoccupied buildings is not possible, because of the smahus. Värme och vitvaror-fullskaleförsök, Rapport TN:42, The
often high correlation between the identified UA and National Swedish Institute for Building Research, Gävle.
gA values. As a result, the confidence intervals on both 11. Lindfors A, Westergren K-E and Lilliestrale M C J, 1998, ‘An
parameters are often (very) large. Moreover, it is often Internet based system suitable for European wide monitoring
not possible to estimate the ventilation losses with a energy use’, in EPIC 98, Lyon, France (19–21 November 1998),
reasonable accuracy. Vol. 2, pp. 606–611.
3. Instead of identifying the individual UA and gA values, it 12. http://www.ericsson.com
13. Westergren K-E, Högberg H and Norlén U, 1996 (revised 1997),
is easier to identify the normalized energy consumption.
An Energy Barometer for Sweden. Pilot Study and Proposal, Build
4. In the case of occupied buildings, there are several
Environment, Royal Institute of Technology, Gävle, Sweden.
additional problems (more important internal gains,
which are often not well known, the influence of user
patterns, the opening of doors and windows, etc.), Save HELP
because of which the uncertainty in the identified 14. Martin S, Wouters P and L’Heureux D, 1996, Evaluation of a
normalized energy consumption substantially increases. simplified method for the energy certification of non-occupied buildings,
In occupied buildings, it does not seem at all possible Final Report, Save HELP Project, Belgian Building Research
to identify the UA and gA values with a reasonable Institute, Brussels.
accuracy if such identification is to be based on the 15. Bloem J J and Martin S, 1998, ‘A pseudo dynamic analysis tool
measured energy use by the building itself. for thermal certification of dwellings’, in EPIC 98, Lyon, France
5. The following trends are important: (19–21 November 1998). Lyon: Ecole Nationale des Travaux
– The relative accuracy is typically worse in Publics de l’Etat, 1998, Vol.2, pp. 403–408.
16. Richalet V, Neirac F P, Tellez F, Marco J and Bloem J J, 1998,
better-insulated buildings.
‘HELP (House Energy Labelling Procedure): Methodology and
– The relative accuracy is typically worse in southern
present results’, in EPIC 98, Lyon, France (19–21 November 1998).
warm climates.
Lyon: Ecole Nationale des Travaux Publics de l’Etat, 1998,
– The relative accuracy is typically worse for short Vol.2, pp. 129–134.
measurement durations.

STEM and PSTAR

REFERENCES 17. Balcomb J, Burch J and Subbarao K, 1992, Short-Term Energy
The university projects & the Energy Barometer Monitoring: An Overview, Golden, CO, Solar Energy Research
Institute.
1. Bostadsdepartementet, 1980, Energispareffekter i Bostadshus där 18. Subbarao K, 1988, PSTAR – Primary and secondary terms analysis
Åtgärder Genomförts med Statligt Energisparstöd. Expertbilaga 5 and renormalization: A unified approach to building energy simulations
till SOU 1980:43 – Program för energihushållning i befintlig and short-term monitoring, Report SERI/TR-254-3175, Golden,
bebyggelse. Ds Bo 1980:8. CO, Solar Energy Research Institute.
2. Elmroth A, Hjalmarsson C, Norlén U, Rolén C, et al., 1989, 19. Subbarao K and Burch J, 1985, Macrodynamic Theory and
Effekter av energisparåtgärder i bostadshus. Rapport R107:1989, Short-Term Measurements for the Thermal Assessment of Buildings,
Byggforskningsrådet, Stockholm, Sweden. Report SERI/TR-253-2854, Golden, CO, Solar Energy Research
3. Westergren K-E, Högberg H and Norlén U, 1998, ‘Monitoring Institute.
energy consumption in single family houses’, in Energy and 20. Subbarao K, 1984, BEVA (Building Element Vector Analysis) – A new
Buildings, 29, 247–257. Hour-by-Hour Building Energy Simulation with System Parameters as
24 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Inputs, Report SERI/TR-254-2195, Golden, CO, Solar Energy 37. Kreider J F and Wang X A, 1991, ‘Artificial neural network
Research Institute. demonstration for automated generation of energy use
21. Burch J D, Subbarao K, Hancock C E and Balcomb J D, predictors for commercial buildings’, in ASHRAE Transactions,
1989, Repeatability and predictive accuracy of the PSTAR short-term 97(2), 775–779.
building monitoring method, Report SERI/TR 254-3606, Golden, 38. Olofsson T, Andersson S and Östin R, 1998, ‘Using CO2
CO, Solar Energy Research Institute. concentrations to predict energy consumption in homes’,
22. Balcomb J D, Burch J D and Subbarao K, 1993, ‘Short-term in Proceedings of the 1998 ACEEE summer study of energy
energy monitoring of residences’, in ASHRAE Transactions, efficiency in Buildings. American Council for Energy-Efficient Economy,
99(2) 935–946. Washington DC. Washington: American Council for an Energy-
23. Subbarao K, Balcomb J D, Burch J D, Hancock C E and Lekov A, Efficient Economy, 1994, Vol. 1, 1211–1222.
1990, ‘Short-term energy monitoring summary of results from
four houses’, in ASHRAE Transactions, 96(1), 1478–1483.
24. Subbarao K, Burch J, Hancock C E and Lekov A, 1988, Short
Other related topics
Term Energy Monitoring (STEM): application of the PSTAR method 39. Pettersen T D, ‘Variation of energy consumption due to
to a residence in Fredericksburg, Virginia, Report SERI/TR3356, climate, building and inhabitants’, in Energy and Buildings, 21,
Golden, CO, Solar Energy Research Institute. 209–218.
25. Burch J, Subbarao K, Lekov A, Warren M and Norford L, ‘Short- 40. Flouquet F, 1992, ‘Local weather correlations and bias in
term energy monitoring in a large commercial building’, in building parameter estimates from energy signature models’,
ASHRAE Transactions, 90, 1459–1477. in Energy and Buildings, 19, 113–123.
26. Balcomb J D, Burch J D, Subbarao K and Hancock C, 1994, 41. Stephenson D G and Millas G P, 1967, ‘Cooling load calculations
‘Short-term energy monitoring for commercial buildings’, in by thermal response factor method’, in ASHRAE Transactions,
Proceedings of ACEEE Summer Study on Energy Efficiency in Buildings, 73, pp. 311–317.
Vol. 5, Commissioning, Operation and Maintenance. Washington:
American Council for an Energy-Efficient Economy, 1994.
27. Byars N, Hancock C, Anderson R and Balcomb J., 1990,
Other measurement techniques and monitoring
‘Experimental evaluation of passive cooling using PSTAR’, in
experiences
15th National Passive Solar Conference, American Solar Energy Society, 42. Burch J D, 1986, ‘Building thermal monitoring methods’, in
19–22 March 1990, Austin, Texas. American Solar Energy Society, Passive Solar Journal, 3(2), 149–177.
1990. 43. Mazzuchi R P, 1992, ‘End-use profile development from whole-
building data combined with intensive short-term monitoring’,
in ASHRAE Transactions, 98(1), 1180–1186.
Neural networks 44. Palmiter L, Toney M and Brown I, 1988, Preliminary evaluation
28. Olofsson T, Andersson S and Östin R, 1998, ‘A method for of two short-term building test method, Ecotope Inc. Seattle.
predicting the annual building heating demand based on 45. Subbarao K et al., 1985, ‘Measurement of effective thermal
limited performance data’, in Energy & Buildings, 28, 101–108. capacitance in buildings’, in Proceedings ASHRAE/DOE/BTECC
29. Kreider J F, Claridge D E, Curtiss P, Dodier R, Haberl J S Conference on Thermal Performance of the Exterior Envelope of
and Krati M, 1995, ‘Building energy use prediction and system Buildings, Clearwater Beach, FL, December 1985. Atlanta: American
identification using recurrent neural networks’, in Journal of Society of Heating, Refrigerating and Air-conditioning
Solar Energy Engineering, 117, 161–166. Engineers.
30. Cammarata G, Fichera A, Marletta L, Micali A and Vagliasindi 46. Flouquet F and Richalet V, 1989, ‘Analyse comparative de
C, 1998, ‘Control of indoor air temperature in buildings based méthodes pour l’évaluation des performances énergétiques
on neural networks’, in EPIC 98, Lyon, France (19–21 November des bâtiments sur site’, in Proceeding Conference on Science and
1998). Lyon: Ecole Nationale des Travaux Publics de l’Etat, Technology at the Service of Architecture, Paris. Dordrecht: Kluwer
1998, Vol.2, pp. 630–634. Academic Publishers.
31. Kreider J F and Haberl J S, 1994, ‘Predicting hourly building 47. Sonderegger R C, 1978, ‘Diagnostic tests determining the
energy use: the great energy predictor shoot-out – overview and thermal response of a house’, in ASHRAE Transactions, Vol. 84
discussion of results’, inASHRAE Transactions, 100, 1104–1118. (1), pp. 691–702.
32. Andersson S, Olofsson T and Östin R, 1996, ‘Predictions of
energy demand in building using neural network techniques on
Calculation methods and parameter estimation
performance data’, in Proceedings of the 4th symposium on Building
Physics in the Nordic countries, Espoo, Finland, Vol. 1, 51–58. 48. Meier A K and Busch J, 1988, ‘Testing the accuracy of
33. Olofsson T and Andersson S, 2000, ‘Long-term energy demand measurement-based building energy model with synthetic data’,
predictions based on short-term measured data’, in Energy and in Energy and Buildings, 12, 77–82.
Buildings, 33 (2), pp. 85–91, 2002. 49. Rabl A, 1988, ‘Parameter estimation in buildings: methods for
34. Olofsson T, Andersson S and Östin R, 1998, ‘Energy load dynamic analysis of measured energy use’, in Journal of Solar
predictions for buildings based on a total demand perspective’, Energy Engineering, 110, pp. 52–66.
in Energy and Buildings, 28(1), 109–116. 50. Subbarao K, Burch J and Jeon H, 1986, Building as a Dynamic
35. Olofsson T and Andersson S, 1999, ‘Analysis of the interaction Calorimeter: Determination of Heating System Efficiency, Summer
between heating and domestic load in occupied single-family Study on Energy-Efficient Buildings 1986, Santa Cruz, California,
buildings’, in Proceedings of the 5th symposium on Building Physics Report SERI TP254-2947, Golden, CO, Solar Energy Research
in the Nordic Countries, in Gothenburg, Sweden, Vol. II, 473–480. Institute.
36. Olofsson T and Andersson S, 2000, ‘Overall heat loss coefficient 51. Busch J F, 1986, ‘A comparison of building thermal perform-
and domestic energy gain factor for single-family buildings’, in ance models using measured data’, in ASHRAE Transactions, 92,
Energy and Buildings, 37 (11), pp. 1019–1026, 2002. Paper No. 3011, pp. 537–549.
 Review of selected theoretical and experimental techniques for energy characterization of buildings 25

52. Jansen J E, 1980, ‘Application of building thermal resistance 64. BIN-IBN, 1987, Calculation of Heat Transmission Coefficients
measurement’, in ASHRAE Transactions, 88(2). Through Walls, NBN B62-002, BIN-IBN, Brussels (in Dutch and
53. Wilson N W, Wagner B S and Colbourne W G, 1985, ‘Equivalent French).
thermal parameters for an occupied gas-heated house’, in 65. BIN-IBN, 1986, Calculation of the Heat Losses of Buildings, NBN
ASHRAE Transactions, 91(2), pp. 1875–1884. B62-003, BIN-IBN, Brussels (in Dutch and French).
54. Agami Reddy T, 1996, ‘Status of residential short-term 66. Belgian Building Research Institute (BBRI), 1997, Study and
performance monitoring methods’ in, Proceedings of the 1996 Monitoring of the PLEIADE Dwelling – Final Report 1994–1996,
International Solar Energy Conference, 359–367. New York: BBRI, Brussels.
American Society of Mechanical Engineers. 67. Somogyi Z (1998), In Situ Evaluation of the Thermal Characteristics
55. MacDonald M and White D, 1990, ‘Explanatory investigation of of Building Components and Buildings Including Comparison with
a new modelling approach and an energy index for multifamily Predicted Performances, PhD thesis, UCL, Louvain-La-Neuve.
buildings in the Pacific Northwest’, in Conference proceedings of 68. Akander J, Jóhannesson G and Norlén U, 2001, The Experimental
ACEEE summer study on energy efficiency in buildings, Pacific Grove, Protocol of Euro-Class – Level B – 2001-01-10, Divison of
CA(USA) 26 Aug–1 Sep 1990. Washington: American Council Building Technology, Department of Building Sciences, KTH,
for an American Efficient Economy, 1990. Stockholm.
56. Subbarao K, Burch J and Handcock C E, How to Accurately 69. Bloomfield D, et al., 1988, An Investigation into Analytical and
Measure the Building Load Coefficient, Report SERI/J254-0405, Empirical Validation Techniques for Dynamic Thermal Models in
Golden, CO, Solar Energy Research Institute. Buildings, Final Grant Report, Science and Engineering Research
57. Subbarao K, 1985, Thermal Parameters for Single and Multizone Council, Swindon (UK).
Buildings and Their Determination from Performance Data, Report 70. Judkoff R and Neymark J, 1995, Building Energy Simulation Test
SERI/TR253-2617, Golden, CO, Solar Energy Research (BESTEST) and Diagnostic Method, IEA Annex 21.
Institute. 71. Vandaele L and Wouters P, 1994, The PASSYS Services: Summary
58. Källblad K and Adamsson B, 1994, The BKL-Method, A Simplified Report, EUR 15113 EN, European Commission, Brussels.
Method to Predict Energy Consumption in Buildings, Document 72. Bloem J J (ed.), 1994, System Identification Applied to Building
D8:1984, Swedish Council for Building Research, Stockholm, Performance Data, EUR 15885 EN, European Commission,
Sweden. Brussels.
59. Kaplan M, Jones B and Jansen J, 1990, ‘Model calibration 73. Martin S, Wouters P and L’Heureux D, 1996, Evaluation
with monitored end-use data’, in 1990 ACEEE Summer Study on an Application of a Simplified Method for the Certification of
Energy Efficiency in Buildings, Vol. 10, Washington DC, American Non-Occupied Buildings, Report prepared for the EC Joint
Council for an Energy Efficient Economy. Research Center, Ispra under Contract No. EU 11297-95-10
F1EI ISP B.
74. Stymne H and Eliasson A, 1991, ‘A new passive tracer gas
In situ evaluation of UA and gA values
method for ventilation measurements’, in Proceedings of 12th
60. European Committee for Standardisation (CEN), 1997, AIVC Conference – Air Movement and Ventilation Control within
Building Components and Elements – In Situ Measurements of Buildings, Vol. 3, 1–16, Ottawa.
the Surface-to-Surface Thermal Transmittance’, prEN 12494, CEN, 75. Boman C A and Stymne H, 1996, PFT – Measurements of
Brussels. Ventilation in European Dwellings, Report prepared for the EC
61. European Committee for Standardization (CEN), 1996, Thermal Joint Research Center, Ispra under Contract No. EN 9505127 T.
Insulation – Determination of Steady-State Thermal Transmission 76. Helcke G A and Peckham R J, 1991, ‘Energy, the computation
Properties – Calibrated and Guarded Hot Box, EN ISO 8990 of degree-days for use as a climatic severity indicator’, in Eurostat
(European Norm), CEN, Brussels. Energy Statistical Tables, EUR 13697 EN, Commission of the
62. Alexandre J (ed.), 1998, La Maison PLEIADE – Conclusions, European Communities, Luxembourg.
Jambes (in French). 77. European Committee for Standardization (CEN), 1993, Thermal
63. BIN-IBN, 1989, Thermal Insulation of Buildings – Thermal Performance of buildings – Calculation of Energy Use for Heating
Characteristics of Buildings, NBN B62-301, BIN-IBN, Brussels – Residential Buildings, prEN 832 (European Norm), CEN,
(in Dutch and French). Brussels.
26

CHAPTER 2
Experimental methods for the energy
characterization of buildings
JAN AKANDER AND GUÐNI JÓHANNESSON
Division of Building Technology, Department of Building Sciences, KTH, Stockholm

DEFINITIONS Building envelope: All building components that separate

the external and internal environments.
Residential building/apartment: A residential building or Solar aperture: Openings in the building envelope that
apartment may be composed of one or more spaces enclosed permit solar irradiation.
by walls, ceilings and floors. The building must be situated Tap warm water: Tap water that has been heated.
above the ground and be utilized as a dwelling. At least Heat-loss factor: Heat flow coefficient from the heated space
70% of the space within the envelope must be used for this to the external environment (EN ISO 13790).1
purpose. This volume should be heated to at least X ◦ C during Heat gain: Quantity of heat generated within or entering
Y hours of the year. (Values X and Y are to be determined on into the heated space from heat sources other than the
a national level). heating system (EN ISO 13790).1
Energy use: Any form of fuel or electricity, which is actively Internal heat gain: Quantity of heat generated within the
supplied to the building with the intent of conditioning heated space from heat sources other than the heating
the living space, heating water and operating energy systems system.
and household appliances. Energy use is given on an annual Solar gains: Heat gains due to passive solar irradiation.
basis, kWh/(year · building unit). Heat losses: The amount of heat transferred from heated
Global energy use: Total energy use of the whole residen- space to the external environment by transmission and by
tial building or apartment. This value can be actual or ventilation, during a given period of time (EN ISO 13790).1
normalized. External temperature: The temperature of the external air
Specific energy use: Energy use with an explicit function, (EN ISO 13790).1
which, when taken with the other specific energy uses, adds Internal temperature: The arithmetic average of the air
up to the total energy consumption of a whole residential temperature and the mean radiant temperature at the centre
building. The types of specific energy use are space heating, of a room (internal dry resultant) (EN ISO 13790).1
space cooling, tap-water heating, appliances and external Set-point temperature: The design internal temperature
energy. The value can be actual or normalized. (EN ISO 13790).1
Energy conversion system: Any system or device in the Thermal zone: Part of the heated space with a given set-point
building with the purpose of converting fuels or energy into temperature, throughout which the internal temperature
work or heat. is assumed to have negligible spatial variations (EN ISO
Energy conversion system efficiency: The ratio between 13790).1
delivered and supplied energy of an energy conversion Unheated space: A room or enclosure that is not part of the
system. heated space (EN ISO 13790).1
Supplied energy: Energy (or an equivalent quantity) that is Utilization factor: A factor reducing the total monthly or
actively supplied to the conversion systems of the residence. seasonal gains (internal and passive solar) to obtain part of
Supplied energy is often referred to as billed energy. the useful gains (EN ISO 13790).1
Delivered energy: Energy that is delivered into the residential Metabolic heat: Heat dissipated from living beings.
living space through energy conversion systems.
Actual energy use: Energy use that was observed when the
building was utilized in its natural condition for a certain THE EXPERIMENTAL PROTOCOLS
period of time. This chapter presents the experimental protocols that
Normalized energy use: Actual energy use that is adjusted to are to be used to gather information together with the
a set of reference conditions. methods of processing the information to prepare values for
Standard conditions: A set of variables that have pre- normalization prior to the rating procedure. It is important
scribed values. to note that the protocols provide the framework for what
Appliances: Devices used for converting one form of energy is to be taken into consideration within the collection and
or fuel into useful energy (heat or work). processing of energy values. How this is to be done for
 Experimental methods for the energy characterization of buildings 27

individual buildings cannot be specified in the form of the magnitude of the specific energy use. The output of
various case studies: the case combinations would be endless. the protocol is intended to provide an input to the rating
Chapter 5 illustrates eight real-scale applications and how scheme with energy as a basis: the fundamental unit must be
these have been dealt with. The framework leaves certain a thermal unit, i.e. energy expressed in Joules or equivalent
aspects to be determined at a national or regional level. units (kWh for example). The output of the protocol
The motivation for this is that information on building must also be related to the energy that is actually used
technology, end-user behaviour, statistics and informative by the occupants and the building as a whole. This is a
building codes or methods, and customs (traditions) are fundamental criterion within the classification scheme: it
better treated within a local framework. The idea is that the is not a set of pre-determined conditions that will determine
framework can be adapted to suit the national conditions in the outcome because actual observed energy use must be
the most appropriate way. assessed with an experimental protocol. This energy will
then be normalized with reference to the outdoor conditions
that have prevailed and to the building size. Normalization
The need for experimental protocols
is used to make variables usable in various comparison
Rating of buildings by means of energy use is not unusual. scenarios.
There exist a wide range of various rating methods; see From what has just been discussed, it is clear that the
for example the review by Kotsaki and Sourys.2 The rating strategy used for designing the protocol must be based on
methods can be customized for residential or commercial an experimental point of view: the actual energy used over a
building only or they may be applicable to both of these. period of time is the true input to the protocol. The only
The majority of methods make use of descriptive models. observations, or measurements, that are carried out over
In other words, a fundamental feature prior to the rating time without an active monitoring campaign are reflected
procedure is that technical data on the building is assessed in energy bills. These give the values of actual energy use,
and implemented in a model, which, by means of simula- and form the heart of a truly experimental protocol.
tion, gives the thermal performance of the building. The
simulation makes use of some reference climate data The aim of the experimental protocol
and standardized indoor conditions for the purpose of
evaluating the thermal performance of the building, subject The aim of the experimental protocol is:
to these conditions. The simulated performance can then
be compared with that of other buildings, subject to the • to determine actual energy use and identify its
same conditions – how good is the thermal performance of constituents
the considered building, and why is it behaving in this way? • to assess specific energy use, which can be normalized.
The descriptive models allow energy flows to be traced, and
action plans and ratings can be derived from these results. The actual energy use is reflected in energy bills or other
This type of procedure is common, but how well does energy records, which cover a period in the past. The
the descriptive model simulate the true energy use under frequency of the bills is an important factor when it comes to
prevailing conditions? Research indicates that simulated interpreting why the energy is used in a certain way and what
results based on assumed conditions, for example indoor the energy is used for. As long as one is satisfied with knowing
activities and the living patterns of occupants, can give quite how much energy was actively supplied to the building for
poor agreement with the actual energy use. Now, if the a period of one year, bill information is perfectly adequate.
rating of a residence is to be performed, is it reasonable However, this is not the case if the intention is to compare
for it to be based on a simulated building with standard the energy use of one building with that of another. Or
indoor conditions and occupants, a reference climate and if one wishes to examine the influence of weather or the
technical data assessed on the energy conversion systems location of the building – or what the effect is of other
and a building envelope usually based on plan drawings and occupants in the same residence. The actual energy use is
audit? The answer is dependent on the purpose of the rating. primarily dependent on the external climate, which nothing
The purpose of a rating is specified to classify buildings by can control. For this reason, it is convenient to relate energy
their energy use. What affects energy use can be reduced to use to a climate that is known and applies to a set of buildings:
a set of variables and parameters. The dominant ones are as a reference climate. This climate will influence energy use
follows: and, by definition, a building subjected to the reference
climate will lead to an energy use that is called normalized
• the efficiency of energy conversion systems energy use. However, the prevailing internal conditions of
• the characteristics of the building envelope the actual and reference climates are assumed to be the same.
• the external environment It is when determination of normalized energy use is
• the behaviour of the occupants, who explicitly control necessary that information and data from the considered
and affect the internal environment. building, its occupants and the related bills have to be
assessed and analysed. In order to create a method for this
The experimental protocol is intended to serve as a platform procedure, an idealized model including definitions and
for identifying the effect that these variables and parameters assumptions has to be formulated. These are the focus of
give rise to in terms of energy use, and for determining the following sections.
28 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Because normalization will primarily be based on the Supplied energy is often metered, after which the energy
conditions of the external environment, it is climate- is often converted to another form by means of a supplied
dependent energy use that will be a key issue in the energy conversion system installed in the building. The
analysis. Primarily space-heating and space-cooling energies energy conversion system may be a heating system, for
will differ for the prevailing and reference climates. Thus, example where the heat content of natural gas is converted
special attention is given to these two entities, since they to heat. Alternatively, it may be an electric cooker, which
particularly influence the difference between actual and converts electricity to heat, or it may be a heat pump that
normalized energy use. makes use of electricity to generate a cooling thermal load.
What is common to all energy conversion systems is that they
The idealized model of energy flow in have performance indexes, or efficiencies, that indicate how
buildings – definitions well the conversion process takes place.
From the point of view of the protocol, the supplied energy
A model of energy flow in a building is needed as the basis is vital input data, since it provides a basis for estimating
for analysis. The analysis could be performed by following a the energy that is delivered from the energy conversion
quantity of thermal energy ‘from the cradle to the grave’, as systems into the building, the so-called delivered energy.
this quantity changes forms over time. The metaphor ‘from Delivered energy is in fact what is commonly called net
the cradle to the grave’ is intended to describe the supply energy.
of energy from the stage of its being primary energy to the
stage where the energy has been used by an end-consumer DELIVERED ENERGY AND SPECIFIC ENERGY
(Figure 2.1).
Primary energy is a term that involves energy resources The output from supplied-energy conversion systems is
and energy conversion systems on a national level. The thermal energy. This energy is called delivered energy, which
experimental protocol will not explicitly deal with primary is needed for creating the thermal indoor environment, for
energy, although it may provide information that is valuable operating energy conversion systems and for fulfilling the
for analysis of primary energy use. The experimental needs and functions that are required by the occupants.
protocol focuses more on secondary energy, in other To illustrate the difference between supplied energy and
words, the use of energy that is supplied from primary delivered energy, the example of a heat pump can be used.
energy conversion systems. Within this context, this is called Suppose the heat pump has a coefficient of performance
supplied energy. (COP) corresponding to 3. If the supplied energy, in this
case electricity, is 1 kWh, then the heat delivered from the
heat pump is 3 kWh.
SUPPLIED ENERGY
Energy delivered from residential energy conversion
Supplied energy is what can often be called billed energy, systems is expressed in thermal units (kWh) and provides
because energy supplier companies measure the quantity of the basis for comparison scenarios for buildings that are
energy supplied to the end-user (the resident) and bill that going to be rated. The task of the experimental protocol is
quantity. The protocol aims to analyse energy use from the to determine delivered energy from the energy conversion
point at which supplied energy is metered to the point where systems in the actual building and to normalize this
energy exits the residence. quantity.

Qmetabolic

Qsolar

Primary energy Supplied energy Specific energy Energy losses

Qtype1 Qtype1 Qspaceheating

Qtype2 Qtype2 Qspacecooling Qtransmission+vent
Primary-energy
Supplied-energy
conversion Qhot water Residence Qdrainage
conversion
system/fuel (building)
systems Qappliances
refinement
QtypeN QtypeM Qexternal

Billed energy Sub-metered

Metered energy energy*

Figure 2.1 An idealized view of energy flow. The methodology is limited to analysis of the energy flow from the point where energy is billed to end-user
customers (the areas shaded in grey). Q denotes energy flow and the asterisk (∗ ) the areas that are not often metered
 Experimental methods for the energy characterization of buildings 29

Within the frame of the experimental protocol, delivered ENERGY FOR SPACE HEATING
energy is divided into five utilization types. These so-called
The energy required for space heating is heat that is
specific energies are:
delivered from heating units to the heated space during the
heating season. For the period of time considered, delivered
• Energy for space heating: Thermal energy delivered from
energy from the heating units is formulated such that
space-heating units with the purpose of heating the  delivered
delivered
residential space (delivered energy). Q spaceheating = Q heating unit (2.1)
• Energy for space cooling: Thermal energy delivered from The billed energy for the same period corresponds to
space-cooling units with the purpose of cooling the heat supplied to the heating system and to the appliances,
residential space (delivered energy). expressed as follows
• Energy for tap-water heating: Thermal energy delivered
from tap-water heating units to the water, with the supplied
 Q delivered
spaceheating
Q spaceheating = (2.2)
purpose of heating tap water from the temperature it had CF · ηsh
on entering the building to the temperature it has after
Here ηsh depicts the heating season efficiency of each
exiting the heating units (delivered energy).
conversion system (heating system) and CF denotes a
• Energy for internal appliances: Equivalent thermal
conversion factor. The conversion factor relates the thermal
energy delivered from domestic appliances and lighting
unit kWh to the billed energy units. From this point on,
within the residential envelope (delivered energy).
supplied energy will be expressed using the unit kWh.
• Energy for external appliances and spaces: Equivalent
thermal energy delivered from appliances and lighting
ENERGY FOR SPACE COOLING
outside the insulated building envelope. This energy
is provided separately, but not included in the heat The required energy for cooling of spaces is the sum of
balance of the building. In contrast, external energy that specific energies that affect the heat balance of the space.
is delivered from appliances and lighting within the Delivered space cooling energy corresponds to
insulated building envelope, but not within the space of delivered
 delivered
Q spacecooling = Q cooling unit (2.3)
the considered apartment, is included in the rating of the
building. The billed energy for the same period corresponds to the
heat supplied to the heating system and to the appliances,
There are two specific energies that are not accounted as expressed as follows
being delivered energy from residential conversion systems.
These are commonly part of the heat gain, which in essence supplied
 Q delivered
spacecooling
Q spacecooling = (2.4)
is the heat delivered into the living space, but which is CF · ηsc
not controlled, or is subject to limited control, in terms Here ηsc depicts the seasonal efficiency of each conversion
of thermal climate. This energy is not purchased, and is system and CF denotes a conversion factor. The conversion
represented as: factor relates the thermal unit kWh to the billed energy units.

• Energy from solar irradiation: The net solar energy that is ENERGY FOR TAP-WATER HEATING
transmitted through apertures into the residential space
The required energy for tap-water heating is the sum of the
and/or for solar heating of tap water.
specific energies that affect the energy balance of the tap
• Energy from metabolism: Heat dissipated from occupants
water. For the period of time considered, the required energy
and other living creatures within the living space of the required
Q hotwater is formulated such that
residence.
required
 
Q hotwater = Q hotwater + Q solarhotwater (2.5)
Various forms of specific energy consumption can consist
of a single fuel source or a combination of different fuel The first term involves active delivery of heat (billed),
sources. Table 2.1 displays possible combinations of fuels whereas the second term involves the case where solar
and their use. collectors are used for tap-water heating.

Table 2.1 Various fuels and forms of specific energy consumption. The symbol (X) implies that the occurrence is very limited
Energy source Space heating Space cooling Water heating Appliances
Natural gas X (X) X X
Electricity X X X X
Fuel oil X X X
Liquefied petroleum gas X X X
Kerosene X X X
Solid fuels X X X
District heating/cooling X (X) X
Solar radiation X X (X)
30 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

The delivered energy is that which is actively delivered to considered. Examples of these are energy for operation
the tap water by means of heating coils, omitting the part of systems, such as central fans and elevators, heating of
that is delivered by solar collectors: common spaces such as staircases, common laundries and
delivered
 lighting in corridors. Although the distribution of this energy
Q hotwater = Q hotwater (2.6)
may be difficult to assess, estimations should be made.
The billed energy for the same period corresponds to the The delivered energy for this purpose must be included
sum of the energy supplied to the tap-water heating units: as a delivered energy to the apartment. The distribution of
supplied
 Q hotwater external energy within the multi-family building envelope is
Q hotwater = (2.7)
CF · ηhw to be determined on a national level. Within this project, the
common external energy is divided among the apartments
Here ηhw depicts the seasonal efficiency of each conversion
on the basis of the part of heated floor area that the
system and CF denotes a conversion factor. The conversion
apartment has in relation to the total heated floor area of
factor relates the thermal unit kWh to the billed energy units.
apartments in the building.
Note that the latter term may derive from a combination of
Delivered energy inside the building envelope is
systems, such as a boiler and a heat pump. 
delivered
Q external = Q external devices and spaces (2.13)
ENERGY FOR APPLIANCES AND LIGHTING

The required energy for appliances and lighting is the sum ENERGY FROM METABOLISM
of the energy dissipated from units within the building
envelope of the residence or apartment. For the period of Human beings and warm-blooded pets (occupants) dissipate
time considered, the delivered energy from appliances and heat to their nearest environment. When a heat balance has
lighting corresponds to to be calculated, this heat should be taken into account. The
delivered
 delivered dissipated heat can be estimated to correspond to
Q appliances = Q appliances and lights (2.8)

n

The part of the bills that denotes supplied energy to Q metabolic = i · ti (2.14)
i =1
appliances is
supplied
 Q appliances where i and ti are the dissipated thermal power (W) and the
Q appliances = (2.9) period of presence (h) of occupant i, and n is the number of
CF · ηap
occupants. The period of presence is obtained from an audit
Here ηap depicts the seasonal efficiency of each conversion
record.
system and CF denotes a conversion factor. The conversion
This energy is not included in the rating entities, but
factor relates the thermal unit kWh to the billed energy units.
will affect the magnitude of the energies obtained from the
Most household appliances, lighting and other appliances
assessment procedures.
that are used for operating systems in the building make use
of electricity and some cookers make use of gas. The seasonal
efficiency is therefore very close to unity. However, there are ENERGY FROM SOLAR IRRADIATION
some appliances for which this is not the case. An example of Assessment of solar radiation that is transmitted into the
this is when an open fireplace is used for aesthetic reasons, building requires information on the geometry, orientation,
not for the sake of space heating. The efficiency may in this shading and topography of the glazed envelope elements.
case be close to zero. Solar apertures As (the effective areas of the elements)
represent these parameters. The transmitted solar energy
ENERGY FOR EXTERNAL APPLIANCES
during a period of observation is calculated as follows:
The required energy for external appliances is the sum of the 
 
energy delivered from devices outside the building envelope. Q solarheat = qs,j · As,n,j (2.15)
These are not considered to affect the heat balance of the j n

building. For the period of time considered, the required 2

Here, qs,j (Wh/m ) denotes the total global solar radiation
required
energy Q external is formulated such that on the surface As,n (m2 ) in the orientation j. A requirement
required
 is that data on global solar radiation on surfaces for different
Q external = Q external (2.10)
orientations are available for the period of observation. More
The delivered energy outside the building envelope is information is given in Appendix 2.
delivered This energy is not included in the rating entities, but will
Q external = 0 (2.11)
affect the magnitude of the results of the assessment pro-
Hence, the billed energy corresponds to the supplied energy, cedures. Energy from solar irradiation is primarily used for
which here is normalizing space-heating and space-cooling energies, and
is indeed a part of the heating requirement of the building.
supplied
 Q required
external
Q external = (2.12) Another energy input that solar radiation gives rise to is
CF · ηex heating of tap water. This entity is considered to be a part
In the case of multi-family buildings, there may be a certain of the energy requirement for tap-water heating. However,
energy use within the envelope but outsidethe apartment this energy will not be included in the rating procedure
 Experimental methods for the energy characterization of buildings 31

and will in most cases be considered a part of the specific • Energy supplied to external spaces, for heating and
energy called Q solar , which is omitted as an input to the rating cooling purposes. This is considered to be climate
procedure. Therefore, dependent.

Q solar = Q solarheat + Q solarhotwater (2.16)

Although Q solar is a delivered energy into the residence, it will PROPOSAL OF TWO PROTOCOLS
not have the delivered index included. The access to observations on energy use of a building in
time depends on the periodicity and the quality of the data
Climate-dependent and climate-independent included on bills. Also, there may be energy use that is
variables not billed (non-billed energy use). This may not only be
dependent on traditions and lifestyle in national or regional
When energy use is analysed, it is convenient to make
terms, but it may also differ between energy suppliers and
assumptions on whether or not the specific energy use is climate-
be covered by different agreements between companies and
dependent or climate-independent. A climate-independent
the energy suppliers.
variable will not change in time (or space) as a result of
Apart from the quality of the supplied energy, the data on
changes in the internal and external environments. The
other energy parameters is often limited. In addition to fuel
specific energies that are considered to be climate-independent
use, the data need to perform an energy analysis of a building
are those that are directly affected only by occupancy
are primarily:
behaviour and needs:
• external climate
• Energy supplied to appliances and lighting. Although use
• internal climate
of certain appliances may be more or less seasonally
• occupant behaviour
dependent, such effects will be neglected. For example,
• efficiency and utilization patterns of energy conversion
lighting is in general dependent on solar radiation and
systems.
time of year.
• Energy for tap-water heating. This energy is assumed to In order to perform a more-or-less ‘exact’ analysis of energy
be evenly distributed throughout the year or season. This flow during the current year and a reference year, a whole
may not be completely true because, for example, year of high-frequency extensive monitoring would have
showering may be more frequent during the summer to be performed. This is not practical. A decision has to
season. The temperature of the incoming tap water may be made, therefore, in which the inaccuracy of the results
also vary seasonally. is weighed against the cost of labour and resources. In
• Metabolic heat dissipated by people and household pets. addition, there is the question of the willingness of the
This is assumed to be the same throughout the year, customer to purchase, not only for a rating score, but also
provided that the occupancy rate within the residence is information on the reasons for the score. This additional
the same. Metabolic heat is somewhat dependent on the service requires more resources and will increase costs. For
temperature of the indoor climate, but within this this reason, two separate protocols have been designed
context will be assumed to be constant. fulfilling the same requirements but with different degrees
• Energy supplied to external devices. This is assumed to of detail. The costs, the resources, the delivered information
be climate-independent on a seasonal basis. Although this (service) and the level of accuracy of the results will allow
is not entirely true, because, for example, of the frequent the customer (occupants) to choose a protocol. The results
use of car heaters during colder periods during the generated by each protocol are compatible in rating since
winter season, this entity is assumed to have a poor they have a similar experimental platform.
correlation with climate. The two protocols are called:

Climate-dependent variables are those that are affected by • the Billed Energy Protocol (BEP)
climate: specific energy use increases when the climate • the Monitored Energy Protocol (MEP).
becomes more severe:

OVERVIEW OF BEP AND MEP

• Space heating. Energy delivered from the heating units of
the residence increases as the climate gets colder. The Assessment of the information that is needed to determine
increase and decrease with climate variations can be the energy use of a residence may require immense resources
regarded as being linearly dependent. and time, thus leading to expensive long-term monitoring
• Space cooling. Energy for space cooling is, unlike space campaigns. For this reason, the two protocols BEP and
heating, non-linear. However, the energy used for cooling MEP are available. What distinguishes these two is the level
does increase when the outdoor climate becomes warmer of detail in data and information, both in the input and
and with increased solar radiation. in the output. The person who desires to rate a building
• Solar energy into the residential space. This is makes a choice, which will depend on what input data
climate-dependent, but can to some extent be controlled and information is available, whether procedure criteria
by means of shading devices. are fulfilled, what types of results are desired and how much
32 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

AUDIT PROTOCOL MEASUREMENTS

Administrative info
Housing characteristics
Household characteristics BILLS
Space heating
Air conditioning
Water heating
Internal appliances
External appliances
Fuel bills info

NON-MONITORING ASSESSMENT
ASSESSMENT PROCEDURES
PROCEDURES Fuel conversion factors
Fuel conversion factors System efficiency
System efficiency BEP MEP Hot-water consumption
Hot-water consumption Etc.
Etc.
Actual climate data
Actual climate data Reference climate data
Reference climate data

OUTPUT
Actual annual energy use: Supplied and delivered
Normalised annual energy use: Supplied and delivered
Technical information on the building

Figure 2.2 The two protocols BEP and MEP, which can be used for assessing the energy supplied and delivered and the technical information. Each
protocol requires a different amount of input data amount and different assessment methods. The input sources are shaded

they are willing to pay for this service. The two protocols the boiler efficiency as a function of fabrication, type
are shown in Figure 2.2. and age.
The simpler protocol, BEP, is conducted within four man- The protocols are summarized in Tables 2.2 and 2.3, which
hours per residence, provided that the input data fulfil give an overview of the general issues for each procedure.
several criteria. The output consists of the actual and the For more detail, please refer to the following sections in this
normalized annual supplied and delivered specific energy chapter, which give more detail on each procedure.
use. Normalization of energy use is only performed with
regard to the outdoor climate, because indoor temperatures
are unknown. To decrease the cost of the rating process, BILLED ENERGY PROTOCOL (BEP)
the only measurements that will be made will be reading
The Billed Energy Protocol (BEP) is the simpler of the two
meters that already are installed in the building. The fact that
experimental protocols. Relying heavily on bill information
measurements are not conducted will lead to an increased
and earlier records, this protocol receives its name based
uncertainty in the results.
on the source of the most important information that is
MEP makes use of bill records, but is more flexible since
assessed. Also using information obtained from an audit at
the protocol involves measurements of climate variables and
the residence, this method does not explicitly make use of
sub-metering within the building to enhance the accuracy
any monitoring campaigns. The protocol is to be performed
of the input data. MEP is also recommended in the case
at a minimum cost. However, in order for the outcome of
where the criteria for BEP prevent application of the simpler
the protocol – the results of energy use – to be adequate
protocol. MEP output will in most cases be more reliable
in terms of accuracy, there are several requirements that
than that from BEP. Moreover, normalization of energy use
must be fulfilled in order for BEP to be executed. If these
is performed with the reference outdoor climate and the
requirements are not met, the other protocol (MEP) should
indoor climate taken into consideration. The monitoring
be applied.
period is at least two weeks long, although from an energy
point of view two weeks is considered as a short-term
measurement. MEP may involve a concentrated monitoring
scheme, lasting for over 10 weeks. The duration of the
Strategy
monitoring depends on which service, aside from rating, is BEP relies on information from bills and the audit as input.
purchased. Technical information from MEP is important The procedures within BEP that are used to derive specific
to document, because this information can be used to energy will, in cases where there may be lack of information,
establish default values for BEP. An example of this is make use of estimations or default values. Because it
 Experimental methods for the energy characterization of buildings 33

Table 2.2 An outline of BEP and MEP

Protocol BEP MEP
Actions Audit at site Audit at site
Use of energy bills Use of energy bills
Use of monitored data
Measurements Readings of existing gauges during Metering of supplied or delivered energy
the audit occasion
Measurement of energy system efficiency
Indoor and outdoor temperature measurements
Optional solar radiation measurements
Optional heat loss factor measurement
Optional measurement of ventilation rates, air-tightness
Duration of audit 1 – max 4 hours 2 – max 8 hours
Duration of measurements None A minimum of 2 weeks – about 10 weeks, depending on building type
and purpose

Table 2.3 Input and output variables for the two protocols
Protocol BEP MEP
1. Supplied energy Billed quantities Billed quantities
Measured (min 2 weeks)
2. Efficiencies Default values or from previous tuning records Measured
3. Specific energies Estimated annual use Measured (min 2 weeks)
Estimated annual use
4. Delivered energy Estimated on basis of 1, 2 and 3 above Estimated on basis of 1, 2 and 3 above
5. Outdoor temperature Closest climate station Measured (min 2 weeks)
Climate station for annual data
6. Indoor temperature No Measured (min 2 weeks)
Mean value assumed same
for season of measurement
7. Solar radiation Closest climate station Measured (min 2 weeks)∗
Climate station for annual data
8. Normalization of outdoor climate Necessary Necessary
9. Normalization of outdoor and indoor climate No Necessary
10. Determination of residential heat loss factor No Optional†
11. More detailed technical building data No Optional‡
∗ Optional, depending on whether or not solar radiation has a large influence on the heat balance of the building during the monitoring period.
† If possible, depending on the data quality and the season during which measurements were performed, and whether the customer desires this service.
‡ Extra services such as measurement of rates of ventilation, air-tightness of the envelope, indoor air quality, thermal bridges, U values for specific

envelope components and lighting characteristics.

depends on the quality of the bills, the billing periodicity, following:

the availability of conversion system tuning records, sub- • Pre-arrangements: The auditor informs the occupants
metered values and reliability based on information supplied about the experimental protocol, how it works,
by the end-user, the outcome of the procedures should limitations etc. A quick check is made on whether or not
always be considered to be an estimation. BEP does not the residence fulfils the BEP requirement list.
strive to evaluate the magnitude of the inaccuracy that • Collection of billed data: Series of bills are collected and
the estimations involve: this inaccuracy will not only vary the quality is evaluated.
depending on which building is considered; it will also • Collection of non-billed data: Series of non-billed
depend on the systems, end-user behaviour, bill quality, quantities are collected (on the basis of information
availability of climate data and the judgements of the supplied by the end-user or a third party).
auditor. It will also depend on default values or national • The audit: General information is assessed for the
standard calculation or statistical methods, which are used building. For this purpose, a checklist is provided for the
as estimation procedures when data for the considered auditor. The general information covers aspects such as:
building is not assessable. – administrative and technical data on the building itself
A step-by-step illustration is given below, summing up – technical data on the energy conversion systems and
the major steps involved in BEP. The steps are described energy distribution within the residence
in more detail in succeeding sections. These involve the – end-user behaviour.
34 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

• Analysis of bills: The analysis will take into consideration: If the protocol only dealt with supplied energy, the
– energy supplied to each energy conversion system summing of bills would give adequate information on energy
– an estimate of the delivered energy from each energy use. However, more information and procedures have to
conversion system be applied if the output of the protocol is to be delivered
– an allocation of each delivered energy to the correct energy with allocation into the various specific energies.
specific energy, special attention being focused on the Furthermore, if these values are to be normalized, more
specific energies for space heating and space input is required.
cooling. For the protocol to be affordable, both in time and
Making assumptions on, for example, seasonal energy resources, it has to be simple, but it must also be
use can ease the analysis of bills. weighed against result inaccuracy. This means that the
• Optional services: If the quality of the input data is procedures within BEP may rely quite heavily on estimations
sufficient, other informative services may be provided, – certainly for allocation of each of the delivered energies
but are not required, within BEP. Such a service can be to a specific energy category. Moreover, the efficiency of
to estimate the heat loss factor of the building. energy conversion systems may not have been measured –
• Results: The outcome of BEP is the following: alternative methods will have to be used in order to proceed
– an annual record of the supplied energy with the assessment process.
– an annual record of the estimated values for the five In Table 2.4, the limitations of BEP are listed, including
specific energy types. some comments as to the reasons for the limitations.
It is necessary to specify how the values have been
obtained and that the calculations are described in the
accompanying list of footnotes.
Conditions for the application of BEP
As a result of the design of BEP, it is necessary
Pros and cons
that certain conditions be fulfilled. These conditions
BEP is completely based on information that can be acquired are necessary to limit the uncertainty in part of the
from previous bills and from an audit that does not include input data, which is introduced into the procedures.
monitoring. BEP is experimental in the sense that billed Should these requirements not be fulfilled, it will not
quantities are often measured and descriptive calculation be possible to obtain reliable results. Rating will not be
models are not used. In fact, the major input data is energy possible according to BEP. The alternative is to proceed
supplied during the period considered. with MEP.

Table 2.4 The limitations of BEP

Aspect Assessment Comment
Indoor climate Not considered The occupant supplies information on indoor temperatures, but this is considered to be
somewhat unreliable. Monitoring is not performed because of time limitations. Set-point
temperatures, if possible, are assessed during the audit (monitoring is done in MEP)
Temporary value from
the audit or information
supplied by the occupant
External climate Estimated Data from the nearest climate station can be used
Not measured Discrepancies in micro- and macroclimate are not considered to be greater than
inaccuracies in other steps (local monitoring is done in MEP)
Energy conversion Estimated Time-consuming and costly
system efficiencies Not measured Default values, older tuning records or national procedures can be used (measurements
made in MEP)
Sub-metering of Estimated Time-consuming and costly
specific energy Not measured The largest amount of estimations are made for multipurpose fuel use. The allocation of
delivered energy to specific energy requires estimations. Estimations can be made using
default values, statistics or prescribed codes. These have to be determined on a national
level, to suit the framework for country-specific constraints. The use of estimation
procedures introduces errors into the allocation, which will affect later calculations on
normalization (sub-metering is done in MEP)
Solar energy and Estimated Solar energy and metabolic heat can be calculated from relevant assessed values (solar
metabolic heat apertures and occupant presence), but with large uncertainty. Solar energy will require
computational procedures (the same applies in MEP)
Non-billed supplied End-user estimation Non-billed supplied energy information has to be supplied by the end-user or a third party.
energy The reliability of this information cannot be verified. If the annual value exceeds the global
annual energy by 30%, MEP must be used (short-period measurement is used in MEP)
 Experimental methods for the energy characterization of buildings 35

THE REQUIREMENTS CONCERNING THE RESIDENCE AND codes should contain information on devices, types,
THE OCCUPANTS rated power/output, age, performance indexes and
efficiencies, performance degradation over time, etc.
• The residence must fulfil the definition of being a
Values taken from these lists or codes are to be used as
residence.
default values.
• The building should not have undergone retrofitting
• A regional or national statistical correlation must exist
or extension or had any changes made to the energy
between tap-water consumption and hot-water
systems during the period of time that is covered by the
consumption, or information must be available on
set of bills being used (primarily the last year). The same
hot-water use as a function of the number of occupants
applies to the number of occupants, their behaviour and
or the building space. These values can be used as
their utilization patterns and the rate of
defaults for assessing energy for tap-water heating.
occupancy.
National methods or codes can be applied.

THE REQUIREMENTS CONCERNING THE BILLED AND RECORDED

ENERGY CONSUMPTION Step-by-step description of BEP
• There must be complete sets of bills or readings from PRE-ARRANGEMENTS
gauges that correspond to the annual total energy When the occupants of a residence have expressed interest in
consumption. The bills have at least to include the classification process, the auditor gives them information
information on the fuel type, the quantity of fuel and about the ‘Classification’, how it works, its limitations, and
the time of delivery, as well as whether or not the billed what is required etc. A quick check is made on whether or
values are estimations/predictions on the part of the not the residence fulfils the BEP requirement list.
energy supplier.
• At least two bills for each fuel have to be assessed during
the year. If this is not possible, readings of gauges should COLLECTION OF DATA ON THE SUPPLIED FUELS
be made between the seasons. Values read during the The auditor collects a series of bills. The bills can
audit are considered to be billed values. be obtained from occupants or/and energy suppliers.
• If the bills are based on estimations by the energy Preferably, the bills should cover the most recent year/years
suppliers, it is necessary that the billed estimated values because building performance or user behaviour may
and the readings of values do not radically differ at the change over time.
time of audit. The maximum deviation allowed is 30% Series of non-billed quantities are also collected and the
on an annual basis. end-user or a third party provides information about these.
• For fuels used, but not billed, it is necessary to have It is preferable that this information can be verified. In the
recorded data on an annual basis, as for ordinary bills. worst case, the end-user will have to make an estimation
If this information cannot be verified, the procedure of the quantity used. The estimation can either be made
cannot be executed unless this non-billed fuel comprises on a seasonal basis or by providing information on the
less than 30% of the total energy consumption. frequency of use and how much fuel is consumed on
• The collection of reliable data may include a period of each occasion.
more than a year if relevant input values are to be obtained. The collected data is entered into Form table 1:
For example, fuel oil may only be delivered once a year.
• Conversion factors from fuel types to the unit Wh must
be available. The energy content of each fuel type is • Row 1: The type of fuel is noted in the first row. The
determined at a national level. number of columns depends on the number of fuels used.
• Row 2: For billed fuels, whether the billed quantity is
estimated or measured is entered.
THE REQUIREMENTS CONCERNING AUDIT AND
• Row 3: The conversion factor for heat content per unit
CLIMATE INFORMATION
fuel must be listed, including the conversion units.
• The energy auditor must have inspected the site, accessed • Rows 4–: The period and the billed quantity are entered
all premises, read all relevant gauges and completed an with the thermal unit kWh.
accurate audit form. It is assumed that the residence • Row n: Non-billed fuel quantities are entered. Since
representative provides ‘true’ behaviour and utilization these are not billed, they are considered to be estimates
patterns. for the corresponding periods.
• Reliable climate data must exist for the area in which the
building is situated and must be obtainable from the
THE AUDIT
closest climate station. Reference climate data must exist
for all regions for calculation of the Climate Severity The audit provides an opportunity for the auditor to
Index or other climate data that influences the heat physically assess general information on the building. For
balance of the building. this purpose, an audit form is provided for the auditor. An
• It must be possible to access databanks or national codes audit form is proposed and described in Appendix 1, but
for energy conversion system efficiencies. These lists or it is recommended that national audit forms be created to
36 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Form table 1 Supplied energy into consideration:

SUPPLIED ENERGY Type 1 Type 2 Type 3
Estimated/Measured • energy supplied to each energy conversion system
Conversion factor (Form table 1)
Billed: period 1 • estimated delivered energy from each energy conversion
Billed: period 2 system (Form table 2)
Billed: period 3 • Allocation of each delivered energy to the suitable
... specific energy, special attention being given to the
Non-billed: period specific energies for space heating and space cooling
SUM (kWh/year) (Form table 3).

make the inquiry more targeted. The general information Form table 3 is composed of non-shaded blocks rep-
covers aspects such as: resenting delivered energy from each energy conversion
system considered. The sum of elements in each row
(Period . . .) must correspond to the fuel supplied to the
• technical data on the building
system multiplied by the system efficiency. Each element is
• technical data on the energy conversion systems and
an allocation with respect to the specific energy to which
the energy distribution within the residence
the delivery from the system was made. The lowest row for
• end-user behaviour.
each non-shaded block (SUM) shows on an annual basis the
system’s contribution to the specific energy.
The information from the audit is complementary to the bill
Of great importance are the shaded rows (Method). The
information and can be used for the purpose of analysing
footnote is a number that points to a footnote, which gives
the energy use. An important issue is assessment of the types
more detail and accounts for:
of energy conversion system that are present in the building,
and assessment of the efficiency of these systems. These are
entered in Form table 2. • what assumptions were made (seasonal calculations,
climate dependency, etc.)
• which methods were used (national codes, statistical
• Column 1: A description of the energy conversion system
relationships, etc.)
is entered here. Some examples are:
• how the values were obtained/derived (motivations,
– boiler for tap water and space heating (mark, age,
calculations, equations, etc.).
rate power, etc.)
– lighting, household appliances and services
– stoves (type, mark, age, rate power, etc.) The footnotes are lists that are separately attached to the
– portable electrical radiator (type, mark, age, table. An important feature is that the footnote information
rate power, etc.) shows calculations and references to methods. The results
• Column 2: Fuel type used by each energy conversion and calculation procedures are thus transparent.
system. Some systems may use two fuel types and the The last row (SUM (all)) is the sum of all the system
estimated proportions should be noted. sums in each column. These values represent annual specific
• Columns 3 and 4: Winter and summer efficiencies for energies, which will form the inputs to the normalization
each system are entered here. and rating procedures.
• Column 5: Are the efficiencies estimated E, measured M
or calculated C according to some norm? One of the
OPTIONAL SERVICES
alternatives is entered with a comment on which norm
was applied. Estimated values should indicate whether If the quality of the input data is sufficient, other informative
or not these are national recommended values or services may be provided, but these are not required within
manufacturer values. BEP. Such a service can be to estimate the heat loss factor
of the building (see sections on MEP). In certain countries
such options may be required. For example, in Germany
ANALYSIS OF BILLS
and Switzerland, where energy-use monitoring for individual
Analysis of bills can be eased by making assumptions about, apartments is required by legislation, the data quality will
for example, seasonal energy use. The analysis will take have to be high enough for this option to be mandatory.

Form table 2 List of energy conversion systems, their efficiencies and how these were assessed
ENERGY CONVERSION SYSTEM Fuel type Winter efficiency Summer efficiency Estimated/Measured/Calculated
System 1 Type 1
System 2 Type 2
System 3 Type 3
...
 Experimental methods for the energy characterization of buildings 37

Form table 3 Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1 Footnote 2 Footnote 3 Footnote 4 Footnote 5
SYSTEM 1 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
Period 1
Period 2
Period 3
...
SUM
Method Footnote 6 Footnote 7 Footnote 8 Footnote 9 Footnote 10
SYSTEM 2 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
Period 1
Period 2
Period 3
...
SUM
Method Footnote 11 Footnote 12 Footnote 13 Footnote 14 Footnote 15
SYSTEM 3 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
Period 1
Period 2
Period 3
...
SUM
SUM (all) kWh/year

RESULTS with weighting factors. A convenient weighting factor is the

use of heating or cooling degree-days over the period. Energy
The outcome of BEP is the following:
use is assumed to be proportional to the degree-days of the
period considered.
• an annual record of supplied energy (see Form table 1)
• an annual record of estimated values for the five
specific energy types (see Form table 3). It is necessary TAP HOT WATER
that how the values have been derived is clear in the
presentation. It is seldom that energy for heating tap water is measured.
The reasons are to a great extent that sub-metering is never
carried out:
Processing of data
The analysis of bills is the heart of BEP. Though this is a
• The heat for tap water may be integrated with the boiler
critical phase, it is difficult to provide guidelines to cover
for space heating.
every situation that can arise within the whole building
• The quantity of water supplied to the residence is
stock. The following sections are therefore more descriptive
measured, but not its distribution into the hot and cold
in manner and highlight some of the issues and problems
water circuits.
that may be encountered.
• Solar collectors, if present, deliver heat on the basis of
the outdoor climate.
FREQUENCY OF BILLS

Bills may not be periodically distributed in time. When There are basically two ways of estimating heat delivered for
examining a year, it is important that the bills are compiled water heating purposes.
so that their values coincide within the same year and each
season. For this reason, interpolation between the time of
bills is needed to find how much energy was consumed • determining the energy supplied to the tap-water heating
during the period of interest. system
There are two types of energy use. The first is the climate- • determining the energy delivered to the tap water.
independent energy use, which can be considered to be
constant on a daily, weekly or monthly basis. It may be
Supplied energy known
different for different seasons. Climate-independent energy
use is interpolated or extrapolated on the basis of time. The The first of the two ways is to determine how much fuel
second type is climate-dependent energy use. Interpolation has been supplied to the energy conversion system. If this
between various billing periods is conveniently carried out is the case, then the heat delivered for tap-water heating
38 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

corresponds to compares this value to the actual tap-water consumption

 delivered
Q hotwater to ensure that the value obtained for the energy is
supplied
Q hotwater = (2.17) reasonable.
CF · ηhw
• The entire water consumption is multiplied by the
When the supplied fuel quantity is known, multiplication estimated difference in temperature between the waste
by the fuel conversion factor and the seasonal efficiency of water and the water supplied.
the energy conversion system gives the energy delivered for
tap-water heating. For BEP data processing, the conversion It is important within these methods to gather information
factor and the seasonal efficiency make use of default on what influences the patterns of total water use. Total
values or data delivered from energy suppliers or product water use can vary to a great extent if watering gardens is
manufacturers or data from previous tuning records. frequent, if swimming pools or ponds are used or if the water
In the case where a fuel is for multipurpose use, meter, as well as residential consumption, also measures
sub-metering records will be required. They are seldom farming or industrial water consumption.
available in practice.
APPLIANCES AND LIGHTING
Delivered energy known The term ‘appliances’ refers to devices that are used for
The second way of assessing the heat delivered to tap hot converting one form of energy or fuel into useful energy
water requires that the amount of water that has passed (heat or work). For this reason, lighting is, with this
through the system has been recorded and, at the same time, definition, also a part of this specific energy.
that the rise in water temperature has been registered. The Appliances will therefore include refrigerators, freezers,
mathematical expression is clothes washers and dryers, kitchen and household machi-
delivered nery such as cookers, ovens and dishwashers, lights, fans and
Q hotwater = Mhotwater · cp · θ (2.18)
pumps, and stoves and fireplaces that are not explicitly used
where Mhotwater is the mass of water that has passed through for continuous heating or cooling purposes. The energy is
the heating devices during time t, the specific heat capacity delivered within the boundary of the residence (apartment)
of water is cp (4,180 J/kg · K) and θ corresponds to the and will directly affect the heat balance of the residential
mean increase in temperature. space.
In the case where the mass flow of water has been recorded, A common characteristic is that there are no appliance
but no temperature rise is available, the temperature rise thermostats that control heat (or cooling) dissipation.
may be estimated on the basis of the set-point temperature These devices will continue to dissipate heat even though
of the water-heating device or from the accumulation tank the space temperature exceeds the set-point temperature of
and the temperature of the supplied tap water. The supply the heating system, often resulting in space temperatures
temperature of the tap water can be assumed to be the mean being higher than the set-point temperature of the heating
ground temperature of the season. system.
Sub-metering of this type is not common, but may be Most appliances used for household functions and
present in multi-family buildings with individual monitoring operation of the residence use electricity, apart from
in apartments. cookers, ovens, clothes dryers and lighting, which may use
gas. For the case where only electricity is used by appliances,
Supplied and delivered energies unknown bill analysis will give the delivered energy.
When sub-metered values of fuel and hot-water consump-
STOVES, FIREPLACES AND OTHER HEATING UNITS WITHOUT
tion are not available, estimation of delivered energy must
THERMOSTAT CONTROLS
be applied on the basis of statistical correlations. These may
be derived on a national or regional basis. Examples of There are also appliances that do not explicitly use electricity.
these are: Examples are stoves and fireplaces that are used more for
aesthetic or ‘comfort’ reasons rather than for continuous
• The ratio between the consumption of domestic hot space heating. Stoves may use electricity, wood, pellets, coal,
water and total water consumption should be estimated kerosene, gas or other fuels. If the quantity of fuel is not
by statistical means. With the use of the system set-point billed or recorded, the end-user will be required to make
temperature and the supply temperature of the tap water, estimations of fuel consumption. This is done in one of the
the energy delivered to water heating is estimated with two following ways:
Equation 2.18.
• Hot water consumption can be estimated, based on one • The end-user is required to give information on the fuel
variable or a set of variables, such as the number of type used, the frequency of use of the appliance per week.
occupants or the size of the household. It is important The quantity of fuel used per unit time is either estimated
that the analysis compares this value with the actual by the end-user or is given a national default value. The
tap-water consumption to ensure that the value obtained default value may be based on the type and age of the
for the energy is reasonable. appliance in question.
• Tap-water heating energy for a standard family or a single • The end-user makes estimations on the seasonal
occupant can be used. It is important that the analysis consumption of each type of fuel.
 Experimental methods for the energy characterization of buildings 39

Energy delivered to the space is estimated by means of entity. Depending on how external energy is used, two
methods are available for this estimation.
supplied
 delivered
Q appliances
Q appliances = (2.19)
CF · ηap
Estimation of energy use for a detached non-residential
Default values are used for conversion factors and the heated space
efficiency of the appliance. If the efficiency of the appliance
Estimation of energy use for a detached non-residential
is available, either from the product manufacturer or from
heated space can be done either by estimating how much
previous tuning records, then these values should be used.
energy is supplied into the space or by estimation of the
If energy that is delivered from an appliance exceeds 30% of
heat losses.
the total delivered space heating, then MEP should be applied. The
Estimating the heat-loss factor of the space involves
uncertainty in end-user estimations is the underlying reason for this.
estimation of the heat losses. The auditor assesses the areas
and estimates the U values of the envelope component. The
EXTERNAL ENERGY USE rate of ventilation is guessed. The internal temperature can
External energy use is different from the other specific be measured during the audit and this value can be assumed
energies for three reasons: to be constant during the season. Otherwise, the set-point
temperature is read, or the residence representative can
• External energy use is divided into two categories. These provide information on the mean temperature of that space.
are (a) energy use in the external environment and The heating degree-day concept is used such that
(b) energy use within the envelope of the building,  n·V·ρ·c
but outside the considered space or apartment. Uj · Aj + 3600
HDD
supplied j
• External energy use does not explicitly affect the heat Q external = (2.20)
ηex · CF
balance of the considered residence.
• External energy use can be either climate-dependent where
Uj · Aj is the U · A value for the envelope component
or climate-independent.
j (W/K),
n is the rate of air changes in volume V (h−1 ),
External energy use in the external environment
ρ · c is the product of the density (kg/m3 ) and
Energy is often used in the external environment, especially the specific heat capacity ( J/kg · K) of air,
in the case where the property extends farther than the HDD is the number of degree-days for the considered
building envelope. Energy used for this purpose may be period (K · days).
climate-dependent or climate-independent. What is meant
by the external environment is that the energy use is outside Estimation of energy use of external devices/appliances
the residence envelope. Examples of these are the following:
In the case of devices or appliances in the external
environment, and if the energy use is not sub-metered, this
• a detached heated garage and/or car heaters
energy use has to be estimated. This estimation is preferably
(climate/season-dependent)
made for each season. The auditor is then required to:
• external lighting (seasonal dependency with the latitude
of the building site and the season)
• assess the rated power for the device, either by reading
• heaters for sun-courts, greenhouses, porches, roof
labels or manufacturer information, or measuring at
balconies etc. or other detached spaces
site – default values can be used for specific products
(climate-dependent)
• be informed by the residential representative on the
• pool heaters and pumps
frequency of use or the duration of each use
• grills and gardening appliances (season-dependent)
• make a calculation based on the collected information.
• ground and gutter heating coils (season-dependent).
The energy use corresponds to
External energy use in the external environment is
supplied
considered to be supplied energy because the energy is Q external = rated · t (2.21)
being consumed within the property. At the same time, the where rated (W) is the rated power of the device and t is
energy is not being delivered into the residence and is not the utilization time during the considered season. Caution
affecting the heat balance of the residence. It is not depen- should be taken as to whether or not the rated power refers
dent on the quality and characteristics of the considered to supplied or delivered power.
residence, its envelope and systems. The consequence is By summing the total duration for the whole year and
that this energy should not be a part of the rating scheme. multiplying this value by the rated power, an estimated
supplied
The general application is that Q external is measured or energy use is obtained for each device. It is the responsibility
estimated. Delivered energy will for systems placed in the of the auditor to judge whether or not the energy of a
delivered
external environment be Q external = 0. specific device/appliance can be considered to be negligible.
It is the task of the auditor to estimate the parts of For example, the use of a low-energy light bulb (11 W) for
that supplied energy/fuels that are used in the external external lighting of a porch will not significantly affect the
environment if sub-metering is not present for this specific total energy use of the residence.
40 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

External energy use within the envelope apartments. The considered apartment is thus billed
proportionally to its floor area.
External energy use within the internal environment may
sound cryptic. The meaning of this is that energy is delivered
within the building envelope, but outside the space of the Delivered and supplied energy for seasons
residence. This energy will to some extent influence the heat
The convenience of partitioning annual energy use, de-
balance of the residence.
pending on the frequency of bills, into seasonal categories
Examples are mainly applicable to multi-family buildings
is apparent when the climate-dependent variables for space
and houses that are not detached. These are as follows: delivered delivered
heating and cooling (Q spaceheating and Q spacecooling ) are to be
estimated and normalized in terms of the external climate.
• space heating and cooling of common premises such as
The sum of energy use is on an annual basis formulated
corridors, staircases, hallways, entrances, storage rooms
such that
and garages (climate-dependent)
delivered delivered delivered delivered
• energy for lighting of common premises (certain climate Q annual = Q spaceheating + Q spacecooling + Q hotwater
dependency) delivered delivered
• elevators (climate-independent) + Q appliances + Q external (2.22)
• pumps, fans and other operational machinery functions
that are not explicitly included as internal appliances There may be, in total, four seasons during a year.
(climate-independent) These are the heating season, the cooling season and two
• energy use for operation of common kitchens, intermediate seasons when space heating and cooling is
bathrooms, laundries, clothes-dryer rooms and saunas not applied.
(climate-independent) delivered delivered delivered delivered
Q annual = Q heatingseason + Q coolingseason + Q intermediateseasons
• hot-water use in common premises
(climate-independent). (2.23)

The components of this equation are

Assessment of these entities is extremely complicated unless
delivered delivered delivered delivered delivered
sub-metering is available, which is normally the case. Usually, Q heatingseason = Q spaceheating + Q hotwater + Q appliances + Q external
the costs for these functions are included in residential bills,
and the information on energy use ought to be available (2.24)
through maintenance managers or operators. delivered delivered delivered delivered delivered
The general rule is that energy supplied to external devices Q coolingseason = Q spacecooling + Q hotwater + Q appliances + Q external
supplied
and non-residential spaces is considered as Q external . The (2.25)
energy delivered to these objects is
delivered
 supplied
delivered delivered delivered
Q intermediateseasons = Q hotwater + Q appliances + Q external
delivered
(2.26)
Q external = ηex · CF · Q external

This quantity is included in the energy that serves as The delivered entities are based on estimations of supplied
the basis for the rating procedure. However, although it fuels (billed data and data from the audit) and the
may be climate-dependent to some extent, normalization efficiencies of the conversion systems. The seasons are either
with regard to climate will not be performed. Neither will predefined on a national basis (for example based on the
this energy be taken into account in the normalization regional climate) or on data obtained during the audit. In
of space heating and space cooling of the considered the case where bills overlap the dates that define seasonal
apartment/residence. shift, linear interpolation by time is allowed. The supplied
energy for each season is
• Single-family houses: For single-family houses, external supplied
 supplied  supplied
Q heatingseason = Q spaceheating + Q appliances
energy use is for most cases included in the residential
 supplied  supplied
energy bills. The task of the auditor is to determine what + Q hotwater + Q external (2.27)
part of the energy use goes directly to the external
environment or detached non-residential spaces. This supplied
 supplied
 supplied
Q coolingseason = Q spacecooling + Q appliances
quantity of energy is to be included in the supplied
energy, but will be zero in terms of delivered energy.  supplied  supplied
+ Q hotwater + Q external (2.28)
• Multi-family buildings: Buildings that have common
premises use energy that is not explicitly applicable to
supplied
 supplied
 supplied
 supplied
one user or residence (apartment). A common way of Q intermediateseason = Q appliances + Q hotwater + Q external
financing ‘common/shared energy’ is by summing all
(2.29)
‘common energy use’ and dividing this among the
occupants according to one parameter or a set of
DELIVERED SPACE HEATING
parameters. A model that is widely used for distributing
the common energy costs is to divide the floor area Summing the fuel bill data during the heating season makes
of each apartment by the total floor area of all the estimation of the delivered space-heating energy possible.
 Experimental methods for the energy characterization of buildings 41

The measured or estimated fuel quantities for other specific method is to assess the base load of electrical power during
energies that make use of the fuel type that is used for heating the intermediate periods. Rated values from manufacturers
are subtracted from the sum. This difference, divided or default values for different fabrications or types can also
by the seasonal efficiency of the heating units (default, be used.
estimated or measured value) and the conversion factor of The required energy to obtain the indoor environment
the fuel, gives the estimation of the delivered space heating, and operation of the residence, during the cooling season,
mathematically expressed as corresponds to
required delivered delivered
delivered supplied fuel supplied fuel
Q spaceheating = ηsh · CF (Q heatingseason − Q appliances Q coolingseason = Q spacecooling + Q hotwater + Q solarhotwater
delivered delivered
supplied fuel
− Q hotwater
supplied fuel
− Q external ) (2.30) + Q appliances + Q external (2.34)

This equation assumes that the entities in the brackets are from which the required energy for space cooling
known. In other words, the total amount of fuel that has corresponds to
been supplied during this period has been assessed and the required delivered
Q spacecooling = Q spacecooling (2.35)
parts of the fuel that are distributed for other specific energy
forms have been measured or have been estimated. The implication of this equation is that a house that is
The required energy that compensates for space heat not equipped with cooling units will not have a space-
losses (transmission, ventilation and air leakage), which cooling requirement, since the occupants ‘accept’ the
constitute the internal environment and the operation of thermal conditions of the internal environment. Moreover,
the residence, corresponds to it also indicates that shading devices do not affect energy
required delivered supplied
consumption in the absence of cooling units. However, if
Q spaceheating = Q spaceheating + ηap · CF · Q appliances cooling units are present, shading devices will affect energy
use, reflected within delivered space-cooling energy.
+ Q metabolic + Q solarheat (2.31)

Calculation of the space-heating requirement assumes that DELIVERED DAILY BASE ENERGY USE
heat delivered from the heating units is known or estimated.
Heat dissipated from appliances has to be estimated or The intermediate seasons are when neither space heating
measured. Estimations of heat delivered into the space, nor space cooling is applied. Bills that are from this period
from occupants and solar irradiation, must have been offer valuable data, which to a large extent is climate-
computed. With the underlying assumption that a steady- independent. Assessment of specific energy uses from this
state approach can be applied during the period of time period are what can be considered to be constant on a daily
considered, the required energy for space heating will basis. With the assumption that the base daily consumption is
equal the space heat losses. The space-heating requirement the same for the heating and the cooling seasons, assessment
is used in normalization calculations. of climate-dependent energy use is facilitated.
With these values available, normalization calculations The delivered energy use during this period can be
can commence; see Chapter 3. considered to be the sum of the delivered specific energies,
in an equation expressed as
delivered delivered delivered delivered
DELIVERED SPACE COOLING Q intermediateseasons = Q hotwater + Q appliances + Q external (2.36)
delivered
For the cooling season, the total supplied energy will be the Division of Q intermediateseasons by the duration of the intermediate
sum of all bills for each fuel, given by seasons tint will give the base (daily) power load:
supplied
 supplied  supplied delivered
Q coolingseason = Q spacecooling + Q appliances delivered
Q intermediateseasons
base = (2.37)
 supplied  supplied tint
+ Q hotwater + Q external (2.32)
If this power can be considered to be constant over the year,
The fuel that the cooling units use may also be distributed multiplication by the duration of the heating and cooling
to generate other specific energies. With the fuel con- seasons, respectively, will simplify assessment of the space-
sumption known or appropriately estimated during the cooling and space-heating energies.
cooling season, the delivered space-cooling energy can be Checks should be made, such as:
estimated, such that
delivered supplied fuel supplied fuel • Have the occupants been away from the residence for a
Q spacecooling = ηsc · CF (Q coolingseason − Q appliances
longer period of time during the intermediate seasons
supplied fuel supplied fuel
− Q hotwater − Q external ) (2.33) than in the heating or cooling seasons?
• Are any appliances used differently during the seasons?
Commonly, cooling units are powered with electricity, • Can external appliances be considered to be constant
which is shared primarily with appliances (household over the seasons, or are there additional external
appliances and lighting). This requires estimation of how appliances used during the heating and the cooling
much electricity is used for this purpose. A convenient seasons?
42 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

PROTOCOL FORM Strategy

Municipality House Number ID number MEP can be considered to be a protocol that replaces BEP
in the cases where the requirements of BEP are not fulfilled.
House address:
MEP requires measurements. The measurements will in
Estate property: most cases improve the accuracy of the results. Instead of
the specific energies being estimated, these will be directly
Owner/administrator:
or indirectly measured over a minimum period of two
Owner address (if other than house address): weeks. During this relatively short time period, the order of
magnitude of specific energies will be assessed and indoor
Tenant Name: temperatures will also be measured.
Audit performed (yyyy-mm-dd): The monitoring period is required to be a minimum of
two weeks. However, the duration will to a large extent be
Audit company: dependent on what the monitored data is to be used for.
Auditor name: If the purpose is to determine more accurate values of
specific energy than in the case of BEP, two weeks may
Figure 2.3(a) The protocol form (front page)
be sufficient. However, if it is desired to determine the
heat-loss factor of the building by means of monitoring, the
Corrections to seasonal base loads should be made period may be extended to some 6–10 weeks or longer. If
depending on the answers to the above questions. This can the bills can be considered accurate enough and the billing
involve adding or subtracting average daily power usages frequency sufficient, a two-week measurement of indoor
that are specific for each season. temperatures may be enough to use some other method,
for example the energy signature method or the UA and
THE PROTOCOL FORM gA method, for estimating the heat-loss factor. This assumes
that the indoor temperature during the monitoring period
The form for BEP consists primarily of a front page with was the same as during the billing periods. In other words,
administrative information, followed by three tables and a the service required and the constraints determine the
list of footnotes. length of the monitoring period.
The front page contains administrative information and,
because this type of information varies between countries,
a front page should be produced for each nation. The page Measurement strategy
must contain: The measurements are to be performed for a minimum
period of two weeks. The aim during this period is to
• house or estate identification assess, as part of the internal and external environments:
• house or estate location
• residential address • the indoor temperature
• occupant name • the outdoor temperature
• owner/administrator name • the degree-hours; these can directly be measured or
• contact information (i.e. phone number) calculated from the temperature measurements
• audit date • solar radiation and relative humidity (optional if these
• audit company variables are considered to influence the energy balance
• auditor name of the building during the period in which the
• certification stamp and certification date. measurements are made).

The protocol form is shown as Figure 2.3(a, b, c).

For the monitoring period, the supplied energy must
be recorded. This applies to both billed and non-billed
supplied fuels. Depending on the availability of existing
MONITORED ENERGY PROTOCOL (MEP) metering devices, it may be necessary to install a sensor.
The Monitored Energy Protocol (MEP) is the more detailed Another possibility is for the auditor periodically, during the
protocol of the two. Although MEP takes into consideration monitoring phase, to take readings. This depends on what
monitored data, it also relies on bill information and the data will be used for. In the event that monitored data
earlier records. Complementary information is obtained is required for estimation of the heat-loss factor, readings
from an audit, which is performed when the monitoring have to be performed at least on a weekly basis.
equipment is installed. The protocol is to be performed Another assessment that is required within MEP is to
at a minimum cost, and this is one of the reasons that determine the efficiency of energy conversion systems. This
certain services provided within MEP are left optional. The can be done by two means:
occupant or owner of the residence has the freedom to
choose which service, including the rating process itself, is • direct measurement of the efficiency
to be performed. • an approved equivalent method.
 Experimental methods for the energy characterization of buildings 43

Form table 2 List of energy conversion systems, their efficiencies and how these were assessed
SUPPLIED ENERGY Type 1 Type 2 Type 3
Estimated/Measured
Conversion factor
Billed: period 1
Billed: period 2
Billed: period 3
...
Non-billed: period
SUM (kWh/year)

Form table 1 Supplied energy

ENERGY CONVERSION SYSTEM Fuel type Winter efficiency Summer efficiency Estimated/Measured/Calculated
System 1 Type 1
System 2 Type 2
System 3 Type 3
...
Figure 2.3(b) The protocol form (Form tables 1 and 2)

Form table 3 Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1 Footnote 2 Footnote 3 Footnote 4 Footnote 5
SYSTEM 1 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
Period 1
Period 2
Period 3
...
SUM
Method Footnote 6 Footnote 7 Footnote 8 Footnote 9 Footnote 10
SYSTEM 2 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
Period 1
Period 2
Period 3
...
SUM
Method Footnote 11 Footnote 12 Footnote 13 Footnote 14 Footnote 15
SYSTEM 3 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
Period 1
Period 2
Period 3
...
SUM
SUM (all) kWh/year
FOOTNOTES
1. . . .
2. . . .
3. . . .
Figure 2.3(c) The protocol form (Form table 3)

Approval of equivalent methods has to be on a national fuel). The specific energies are repeated below as a
basis. A possible method is, for example, proposed within reminder:
the Save Belas project.3
A vital part of the measurements is to perform • energy for space heating
sub-metering of specific energies, unless the supplied • energy for space cooling
energy is used only for one specific energy (single-use • energy for heating domestic water
44 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

• internal appliances and lighting performed on rates of ventilation, indoor air quality,
• external appliances/spaces and lighting. air-tightness and thermal bridges, and U values of
components in the building envelope.
The measurements of supplied or delivered energy during • Removal of monitoring equipment: Monitoring
the minimum two-week period are intended to serve as a equipment is removed, and the auditor takes final
basis for finding the distribution of energy for multipurpose readings on installed/existing meters.
energy sources, especially to obtain reliable values for space • Analysis of bills, together with monitored data: The
heating and space cooling. Furthermore, measurements analysis will take into consideration:
can be performed on non-metered energy sources, or on – supplied energy or delivered energy from each energy
energy sources and fuels that are not registered in bills. conversion system
MEP will strive to verify the non-billed information given by – estimated annual delivered energy from each energy
the end-user. conversion system, based on measured entities
For climatic variables, measurements will be performed – the allocation of each delivered energy to the correct
with equipment brought and installed by the auditor. specific energy, special attention being given to the
Specifications on technical data (requirements), placement specific energies for space heating and space
and sampling rates of the various gauges are listed in cooling.
Chapter 2. The auditor is held responsible for the set-up • Results: The outcome of MEP is the following:
and the functioning of the data-acquisition equipment. – an annual record of supplied energy
During the period of measurement, sub-metering of – an annual record of estimated values for the five
supplied energy is to be performed. Because of the wide specific energy types.
range of fuel and system types, and the various combinations This output is directly comparable with that of BEP.
of systems, it is the task of the auditor to determine how the Moreover, the extended services will provide
monitoring is to be performed at a minimum cost. informative results that explain why the energy use is
Since the measurement procedures will require that the way it is.
certain meters and sensors be installed on site, the
auditor will be required to have authorization in certain
Pros and cons
fields or be accompanied by a person with authorization.
Primarily authorization is need within the electrical and MEP consists, from the energy point of view, of a very short
HVAC fields. monitoring period. The objective of the energy measure-
ments is primarily to assess the distribution of delivered
energy among the various specific energies. Secondarily,
PROCEDURE STRATEGY
energy use that is climate-independent is important to
A step-by-step illustration is given below, summing up the identify. The same applies to non-billed fuels. Moreover,
major steps involved in MEP. The steps are described internal temperatures are assessed – both temperature
in more detail in the sections below. They involve the levels and patterns. The effect of the temperatures is the
following: first introduced into the normalization procedures.
The values obtained from the monitoring campaign,
• Pre-arrangements: The auditor informs the occupants together with billed information, contribute to an enhanced
about the experimental protocol, how it works, accuracy in the results when compared with BEP results.
limitations etc. Whereas BEP relies on default-value estimations, MEP
• Collection of billed data: Series of bills are collected. provides actual values. The underlying assumption is that,
• Collection of non-billed data: Series of non-billed for example, the mean daily climate-independent specific
quantities are collected (on the basis of information energy use obtained during the monitoring period is the
supplied by the end-user or a third party). same throughout the season or year considered. MEP
• The audit: General information is assessed for the can be considered to provide BEP with complementary
building. The general information covers aspects such as: information by means of measurements. Nevertheless, within
– technical data on the building itself MEP the bills will still have to be analysed and are
– technical data on the energy conversion systems and the essential energy data within the methodology. The
the energy distribution within the residence only means of not relying on bills would be to deploy
– end-user behaviour. a heavy monitoring scheme lasting for a year or more:
• Installation of necessary gauges: During the audit, the this would provide true annual data on energy use and
auditor has the possibility of investigating what sensors temperatures.
are needed, and where these should be placed. The The extended monitoring period gives more reliable
auditor takes readings on existing meters. values than those of a two-week measurement. In a success-
• Extended services: By extending the measurement ful monitoring campaign, correlation between parameters
period, estimation of the heat-loss factor of the residence and variables may be found. An example is the heat-loss
is possible. This type of data also allows normalization factor of the building. For the purpose of determining
calculations based on normalized indoor conditions. the heat-loss factor of the building, the extended time
Moreover, miscellaneous measurements can be is dependent on which method is being used. Other
 Experimental methods for the energy characterization of buildings 45

Table 2.5 Limitations of MEP

.

Aspect Assessment Comment

Indoor climate Measured∗ Patterns and levels are obtained from the monitoring
Based on information from the occupant† period. These are applicable only to the type of
season when the measurements were made.
Set-point temperatures of the system are, if
possible, assessed during the audit

External climate Measured∗ Data from the nearest climate station can be used.
Climate station data† Discrepancies in micro- and macroclimate can be
assessed during the monitoring period. For annual
calculations, data from the nearest climate station
should be utilized

Energy conversion Measured∗ These are measured, directly or implicitly, during the
system efficiencies monitoring period. Seasonal variations make use of
national default values

Sub-metering of Measured∗ Time-consuming and costly. Specific energy is

specific energy Annual estimation† calculated for the monitoring period, but estimations
have to be made for the rest of the considered
(annual) period. Key values, such as daily
climate-independent energy, are obtained from the
monitoring period. (Estimations can be done using
default values, statistics or prescribed codes. These
are to be determined on a national level, to suit the
framework for country-specific constraints)

Solar energy and Estimated† Can be calculated from relevant assessed values
metabolic heat (solar apertures and occupant presence), but with
large uncertainty. Solar energy will require
computational procedures

Non-billed supplied Measured∗ Non-billed supplied-energy information given by the

energy Annual estimation† end-user can be verified or recalculated on the
basis of measured values. Analysis has to be done
to consider whether or not the consumption is
climate-dependent; this may lead to an extended
monitoring period

Heat-loss factor Measured using environmental temperatures and the heat Usually requires an extended measurement period.
delivered over a period If the billing frequency and quality are adequate, and
corresponding temperatures are known for these
periods, the heat-loss factor may be determined
without the requirement of the extended period
∗ Measured within MEP.
† Estimated on an annual basis.

influencing factors are the building type, the energy systems THE REQUIREMENTS CONCERNING THE BILLED AND RECORDED
and the prevailing internal and external climates. ENERGY CONSUMPTION
The limitations of MEP are listed in Table 2.5,
• A complete series of bills or readings from gauges that
which includes some comments on the reasons for the
correspond to the annual total energy consumption is
limitations.
required. The bills must at least contain information on
the fuel type, the quantity of fuel and the time of delivery,
as well as whether or not the billed values are predictions
Conditions for application of MEP or measured values.
Certain requirements have to be fulfilled for the MEP • Conversion factors from fuel types to the unit Wh are
procedure to be applied. These primarily concern the quality necessary. The specific energy content of each fuel type is
of the series of bills, which within MEP are mandatory data. determined at a national level.
46 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

• The efficiencies of energy conversion systems that could Table 2.6 Room preference for location of temperature gauges, starting
not be measured have to be obtained by using default with the highest priority4
values from national codes or by using other procedures. Gauge 1 Gauge 2
Living room Hall
THE REQUIREMENTS CONCERNING AUDIT AND Hall Bedroom
CLIMATE INFORMATION Bedroom Kitchen
Kitchen
• Reliable climate data must be available from the closest
climate station, and for the two monitoring weeks must
also be measured on site. Reference climate data must the year are most suitable can be done on the basis of old
exist for all regions. energy bills.
• The energy auditor must have inspected the site, accessed From the point of view of the extended monitoring period,
all premises, set up and read all relevant gauges, which primarily aims at determining the heat-loss factor,
measured the relevant variables and accurately filled in the requirements of the method to be used should be
the audit form. It is presumed that the resident followed. For example, the energy signature method gives
representative provides reliable information on more reliable results when the external temperature and
behaviour and utilization patterns. the solar radiation vary significantly during the monitoring
• A criterion that must be fulfilled is that the expected period. In other methods, better results may be obtained if
global energy consumption during the two weeks of the climate conditions are as constant as possible.
monitoring period should correspond to at least 5% of
the annual global energy consumption. CLIMATE GAUGE SPECIFICATION
• When gauges cannot be installed for practical reasons,
default values will have to be used (in the same manner The following information on environments has to be
as for BEP). The limitation on this is that the variable in collected for a period of time:
question must not influence the heat balance of the
building by more than 10%. Otherwise, rating is not • indoor temperature
possible. The default values should be estimated on the • outdoor temperature
basis of reliable codes (determined at a national level) • solar radiation and outdoor relative humidity (optional).
or using alternative methods based on calculations or
statistics. The placement and technical specifications of gauges are
described below for the various types of measurements.
Specifications and requirements
Monitoring of climate variables will require a sampling Indoor temperatures
frequency of half an hour. Detailed information is given in The chosen location for the gauge is one where the
the sections below. temperature is representative of the thermal zone/room.
For measurements of energy flow (global and sub- The room should be one that is frequently used, the doors to
metering), the performance is dependent on the meter the room should normally be open, and the heating/cooling
and gauge types that are actually present, and also on the devices should be set to ‘normal’ room temperature.
possibility of installing new ones. The requirements are set The choice of room, or temperature zone, is to be chosen
as follows: according to the ranking specified in Table 2.6. The ranking
is highest in the first row, with descending preference for
• For electronic monitoring equipment, the sampling rate lower rows.
should be hourly. Some buildings may require more than one temperature
• Older monitoring equipment (already on site) should be gauge. This depends on the number of thermal zones
read at the beginning of the monitoring period and at present in the building. Volumes within a building that
the end of the monitoring period. However, electronic have a temperature difference greater than 2◦ C should be
monitoring, with shorter sampling intervals, is preferred. considered as different thermal zones.
The occupants may perform readings after being Indoor temperature gauges are not allowed to be placed
instructed by the auditor on how to do this.
• on external walls
The inaccuracy of the meters or complementary • closer to external construction joints than 0.5 m
equipment/methods should be no more than ±5%. • closer to external doors or windows than 3 m
The choice of monitoring period during the year should • on internal walls that on the other side have chimneys or
be during one of the seasons when the energy requirement is air ducts
relatively large. A criterion that must be fulfilled is that the expected • on internal walls that on the other side have refrigerators
global energy consumption during the two weeks of the monitoring or freezers
period should correspond to at least 5% of the annual global • on walls that for more than half the day are exposed to
energy consumption. Judging which two consecutive weeks of direct sunshine
 Experimental methods for the energy characterization of buildings 47

• on walls above radiators or other heat sources Table 2.10 Ranges of degree-hours – Outdoor temperature
• on walls above televisions, PCs or other electrical Degree-hours ±9,999◦ C · h
appliances Temperatures At least −28 to +50◦ C
• closer to lamps than 0.5 m
• closer to fireplaces or furnaces than 1 m
Table 2.11 Accuracy, temperature – Outdoor temperature
• so that the gauges are exposed to mechanical stress
Calibration at 0◦ C
• so that the supply air is directed towards the gauge.
Absolute inaccuracy Less than ±0.5◦ C
Discrepancy between indoor and outdoor gauges Less than 0.2◦ C
Technical specifications Discrepancy between outdoor gauges Less than 0.1◦ C
The technical specifications in terms of degree-hours, Long time instability (−30 to +30◦ C) Less than ±0.5◦ C
accuracy and temperature, and sampling rate and data Thermal time constant Around 2 min
collection are given in Tables 2.7, 2.8 and 2.9. Non-linearity (−30 to +30◦ C) Less than ±0.2◦ C

Outdoor temperature Table 2.12 Sampling rate, data collection – Outdoor temperature
Sampling rate 0.5 h
Only one outdoor temperature-measuring device is needed.
Data collection At least a daily mean value of sampled values
For this reason, the location has to be carefully chosen. The
outdoor temperature gauge is not to be placed so that:
Table 2.13 Accuracy – Solar radiation
• direct sunlight influences the gauge; guages are best Nominal spectrum 0.3–3.0 μm
mounted on north façades, preferably sheltered from Non-linearity (<1,000 W/m2 ) ±3%
precipitation by roof eaves Temperature range At least −30 to +60◦ C
• exhaust air is directed towards the gauge
• external doors and open windows are closer than 1 m.
horizontally for global solar radiation assessment. The site
must not be shaded.
Technical specifications
The accuracy specification is given in Table 2.13.
The technical specifications in terms of degree-hours,
accuracy and temperature, and sampling rate and data
collection are given in Tables 2.10, 2.11 and 2.12. Humidity sensor (optional)
Humidity gauges should be placed close to the
Solar radiation (optional for local measurements) outdoor temperature gauge and fulfil the same placement
requirements as the outdoor temperature gauge. This type
Values can be collected from the nearest climate station, of measurement is only of interest if the monitoring
if it is considered that building energy gains due to solar campaign includes energy for space cooling in a humid
radiation are small compared to the heat loss of the building. climate. This is because moisture has an impact on the
In climates with strong solar irradiation, the solar radia- performance of the cooling units.
tion may be measured. The sensor (pyranometer) should
at least fulfil the requirements of ISO 9060 Second Class.5
MEASUREMENT OF SUPPLIED ENERGY
For such measurements, the sensor should be placed
Because of the vast number of types of systems that exist in
the building stock, it is virtually impossible to prescribe how
Table 2.7 Ranges for degree-hours – Indoor temperature
the measurement is to be done for all cases. The auditor
Degree-hours ±9,999◦ C · h
is held responsible for the measurements being carried out
Temperatures At least 0 to +50◦ C
properly and for their being relevant to the results obtained.
The important issues that must be assessed by the
Table 2.8 Accuracy, temperature – Indoor temperature measurement of the supplied energy for a minimum of two
Calibration at 20◦ C weeks are:
Absolute inaccuracy Less than ±0.5◦ C
Discrepancy between indoor and outdoor gauges Less than 0.2◦ C • the total supplied energy during this period
Discrepancy between indoor gauges Less than 0.1◦ C • the distribution of the supplied energy in terms of the
Long time instability (−30 to +30◦ C) Less than ±0.5◦ C specific energies
Thermal time constant Around 2 s • the non-billed energy supplied.
Non-linearity (−30 to +30◦ C) Less than ±0.2◦ C
The gauges and meters that already exist in the building
should be used to the greatest extent possible, after it
Table 2.9 Sampling rate, data collection – Indoor temperature
has been checked that the meters work. The setting up
Sampling rate 0.5 h
of meters should be performed so as to minimize the
Data collection Daily or hourly mean of sampled values
destruction or modification of the systems. In cases where
48 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

metering is not possible, for practical or economical reasons, Oil burners are commonly marked with the following
the assessment methods of BEP can be used. This must be information:
carefully noted on the completed forms.
With electronic monitoring equipment, the sampling rate • manufacturer and model type
should be every half an hour. In some cases, the rate may be • oil flow rate in USgal/h and sometimes litres/h
more intensive, down to seconds. It may not be possible to (litres/h = 3.785 · USgal/h).
equip already existing meters with electronic data acquisi-
tion. For this reason, these types of meters must be read by These flows are based on a pump pressure of 7 bar, an oil
the auditor at the beginning and at the end of the monitor- density of 0.82 kg/litre, an oil viscosity of 3.4 cSt (mm2 /s)
ing period. Residents may read meters manually during the and a temperature of 20◦ C. However, the stated flows may
period, but will not be compelled to do this. Readings made not be accurate. According to manufacturers, the flow may
by residents will require instructions from the auditor. vary by ±5%.
Some systems may be equipped with oil flow meters.
These could be used, but may have inaccuracies of up to
Use of electricity
±10%. The inaccuracies may be greater if contaminants
Most residential premises are equipped with at least one (such as sand) in the oil disturb the function of the flow
electricity meter on site, which may be used. Many brands meter. The oil flow meter can be used as a gauge if the
have built-in possibilities for performing pulse counts. This inaccuracy is less than ±5%.
may require an energized port as part of the monitoring For these purposes, two methods, the nozzle and
equipment. These types of meters commonly monitor the graded cup method, are proposed for determination of oil
total electricity supplied to the building. consumption during the measurement period. The methods
In the case where there is a meter that does not provide require that the run time of the burner be recorded during
the possibility of counting pulses, the readings may have to the monitoring period.
be made manually by the resident or may require comple-
mentary equipment. It is desirable that the auditor interferes The nozzle method
as little as possible with the electrical power supply of the
The nozzle method is the easier one to use. The nozzle is
building, as this requires special authorization.
dismounted so that it can be instrumented with a manometer
The central electrical fuseboard (consumer unit) provides
to measure the pressure of the pump. A correction factor
the possibility of monitoring branch or distribution circuits.
is used to multiply the marked oil flow rate of the burner
Inductive gauges can be used in this case and mounted on the
(at standard conditions as stated above), where this factor is
branch circuits for space-heating systems or devices, space-
defined as
cooling systems or devices and/or domestic water heating.
The electrical consumption for other purposes is assumed to measured pressure
K=
be for household appliances and lighting. 7
For devices that are part of a specific energy other than
The required time is about 30 min and the estimated
household appliances and lighting, and are powered by
inaccuracy around ±5%.6
means of socket outlets on the wall, socket mounted energy
meters can be used. These are convenient if, for example,
portable electrical radiators are used.
The graded cup method
The graded cup method requires that the oil burner be
dismantled. Between the nozzle and the burner body, an
Technical specification
extension hose (produced for this purpose) is connected
The inaccuracy of the meter or the complementary so that the nozzle can be directed into a graded cup. For a
equipment should be no more than ±2%. recorded period (for example, 30 minutes) the amount of
discharged oil is measured.
Use of oil The required time is about 1 h and the estimated accuracy
around ±2.5%.6
Determination of oil consumption during the monitoring
period can make use of oil flow meters that already are
installed in the system. However, the performance of the Comments
oil flow meter should be checked with either the nozzle or The configuration of boilers and burners can vary
the cup method, which are described below. If no oil flow considerably, which requires certain technical skills on
meter exists, one of the two methods below should be used the part of the auditor. However, a specialist from the
to investigate the oil flow through the burner. When this is manufacturer is not needed if the technical instructions on
done, a running-time sensor must be used. Fuel consumption the equipment and procedure are well produced.
during the measurement periods is the run time multiplied When this measurement has been made, it is
by the flow. recommended that there should be no new oil deliveries
The other option is to install a flow meter, which should during the monitoring period. One reason for this is that
be of a type that is compatible with the data acquisition the viscosity of oil can vary between 3.0 and 4.5 cSt. A
equipment. change in viscosity of 1 cSt can lead to a change in flow
 Experimental methods for the energy characterization of buildings 49

corresponding to 2–6%. Another reason is that the heat con- water is commonly not measured in conventional systems
tent of oil may vary by 0.75% between different deliveries. in buildings. The aim of the measurements will therefore
In the case where sensors cannot be installed, readings to a great extent be to determine the part of the fuel
may have to be performed on differences in the oil level consumption that is used for space heating and the part
within the tank. that is used for heating domestic water.
Energy for hot water is to be assessed by measuring one
Technical specification of the following:
The inaccuracy of the meters and the complementary
• the flow of hot water (hot-water consumption) and the
equipment/methods should be no more than ±5%.
temperatures
• the heat supplied to the water.
Use of natural gas
In the first case, a water flow meter has to be installed in
Gas is often transported in permanent pipes, which are the pipe going to the end-user. The energy used can be
equipped with meters (for billing global consumption). estimated to be the amount of hot water that has been used
These meters can either be complemented with a monitoring times the rise in temperature, the density and the specific
device or be read manually. heat capacity. This corresponds to the temperature that
It is unusual for sub-metering to be performed in the the water has on entering the boiler and the temperature
branches of the piping system. This proves to be a problem in the accumulation tank, especially if the water has been
if gas is used for space heating and tap-water heating (a pre-heated by some device.
common combination) and for cooking. For sub-metering, If monitoring possibilities are limited, an approximation
see the section above on Tap hot-water measurements. is to assume that the temperature of the incoming water is
When liquefied petroleum (bottled gas) is used, the the mean annual outdoor temperature, and that the tank
resident is asked to give information on how the bottles temperature is the same as the set-point temperature. This
are used, i.e. for heating or cooking. The simplest means assumes that the domestic water has not been pre-heated.
of measuring liquefied petroleum use is to weigh the bottles In the second case, the heat delivered to the water is easily
that are used during the monitoring period. Another option measured if electrical heating is used. Running-time sensors
is to fit a measurement device to supply tubes. or inductive gauges that are connected to the heating coils
of the system can be used for this. The power of the heating
Technical specification coils is often available.
The inaccuracy of the meters and complementary
equipment/methods should be no more than ±5%. Technical specification
The inaccuracy of the energy metering should be no more
Biomass/wood fuel, coke/coal than ±10%.
Fuel in the form of biomass, wood, coke or coal is
Comments
conveniently weighed or measured (by volume) prior to and
after consumption. The auditor should perform weighing In some cases heating of water taking place in the building is
and/or volumetric measurements at the start and at the not worthwhile measuring. An example is dish- and clothes
end of the monitoring period. Residents/managers may washers that heat water in the washer itself during utilization.
perform daily measurements during the period, but this is The energy source here is electricity and is therefore allo-
not mandatory. cated to the household appliances. However, other models
use water that has already been heated and, in this case, the
energy use will be a part of the domestic water heating.
District heating/cooling
District heating and cooling is often equipped with
energy meters. These are generally based on a form of
Processing of data
flow measurement and a temperature-drop measurement. The results obtained from the monitoring campaign give
Should there be no energy meters present, such a meter indications of the magnitude of the specific energies. Distri-
will have to be installed. bution of energy from each source (fuel type) to a specific
energy has been assessed during the monitoring period.
Technical specification In the cases where BEP values were based on estimations,
The inaccuracy of the energy metering should be no more MEP produces a better estimation on an annual basis.
than ±5%. The description presented below, on how to process data,
depends on what purpose the data serves. Two types of
monitoring schemes can be used:
Tap hot-water measurements
The energy source for heating domestic water is usually • a relatively short monitoring period, in the region of two
the same as, or integrated with, that for space heating weeks
(electricity, gas and oil). Energy for the heating of domestic • a longer monitoring period of around 6–10 weeks.
50 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

THE SHORT MONITORING PERIOD The forms used for BEP are also used for recording the
information. Therefore, as in BEP, one of the forms includes
The short monitoring period, somewhat longer than or equal
the data, and there is also an equivalent form but including
to two weeks, will give limited but yet important information
measured values. The footnotes to the first form indicate
from an energy perspective. Provided that the occupants
what values are used from the measured series.
have lived in a ‘normal’ way, the monitored data will give
A complementary calculation can be carried out if the
an indication of:
quality and frequency of bills is such that these, together
with monitored energy use and indoor temperatures, can
• the size of delivered energies make use of the energy signature method (see below). The
• the size of and the interrelation between specific energies underlying crucial assumption is that the temperature that
• the internal climate (temperature and utilization patterns was monitored during the short-term period is the same
of the energy conversion systems) throughout the considered season. With billed data and
• local external climate variables, as opposed to the climate variables, the use of the energy signature method or
corresponding values from the nearest climate station. a corresponding method can be used to assess the heat-loss
factor of the residence.
For the purpose of assessing values that are to be used for
rating the building, information from the short monitoring
period can be treated in the same fashion in BEP. For this THE EXTENDED MONITORING PERIOD
to work, it is necessary to make the following assumptions The monitoring period must be longer than two weeks if
concerning the acquired data: the criterion of 5% energy use is not fulfilled. However, the
main motivation for extending the monitoring period is to
• Mean base loads measured during the two weeks are determine the quality of the building envelope, which is
constant throughout the season during which reflected in the heat-loss factor.
measurements were conducted. There are several ways of assessing the heat-loss factor, but
• The ratio between the energy for tap-water heating and the costs for these can vary considerably and some methods
the water consumption during the period is almost require the occupants to move out of the residence. The aim
constant during the considered season. of MEP is to assess the heat-loss factor while the occupants
• Measured temperature patterns are similar, throughout continue their normal lives during the monitoring period.
the entire season, to those that were measured.
• Non-billed fuel consumption is summed for the period.
This quantity is divided by the mean temperature Estimation of the heat-loss factor
difference between the monitored internal and external The method used for assessing the heat-loss factor should
climates, and is divided by the number of monitored days. be determined at a national level. The motivation is the
following:
Now, billed information will be used in the same way as
in BEP, with the following steps: • The ways in which building technologies and systems are
used will give different monitoring periods.
• First, the length of the considered season is defined and • The external climate will influence the length of
the corresponding bills are compiled in order. monitoring periods and determine which parameters
• The base loads, multiplied by the duration of each billed should be considered in the fundamental equations.
period, will give the base energy and the delivered energy • It is preferable to use traditional practices and national
that this requires. The delivered energy divided by the norms, since the ways methods are used and the accuracy
system efficiency gives the fuel quantity that is used for of the results is to some extent dependent on experience.
this purpose.
• Base-load fuel quantities are subtracted from the billed Below, the energy signature method is shown to illustrate
quantities. The remainder is climate-dependent: fuel that how such a method may be used.
was used for space heating or space cooling.
• Non-billed fuel consumption, which is considered to be
climate-dependent, is calculated for the corresponding The energy signature method
billed period. Based on the assumption that fuel The energy signature method (ES method) is commonly
consumption ceases (= 0) at the end of the season (the used to estimate energy consumption and can be used
temperature difference between the internal and to determine the heat-loss factor of a building. Based
external environment is known), linear interpolation on energy measurements, the model extrapolates building
with the temperature difference as abscissa is used to energy performance over a longer period than that during
estimate the annual non-billed fuel consumption. which the measurements took place. The method can also
be used to normalize energy consumption.
From this point, BEP procedures can be followed and the Within the framework of this methodology, the ES method
normalization and rating processes can commence. will not use the monitored data from the two weeks. The
 Experimental methods for the energy characterization of buildings 51

length of the monitoring period is too short for the ES where

method to give reliable results. Here, the strategy is, during Ehot water = the energy required for heating domestic
the two weeks, to assess the mean indoor temperature of the hot water
building. This temperature is then assumed to be constant Epersons = the metabolic energy from occupants
throughout the season. By using information from the Ehousehold = the electricity for household appliances
energy bills and accessing climate data for the billed period, The summer factor d has the same formulation as
the ES method can be applied. Equation 2.40, compensated for the length of the summer
season.
The solar aperture f is a parameter that takes into
Description of the model
account solar irradiation within the energy balance of a
The measured energy consumption is statistically regressed building. Moreover, the orientation of the house has to be
against climate data. The model is formulated mathe- taken into consideration. f is given by
matically for an individual building unit7 such that 4
(1 − sd ) · Td · Ad · Id
d =1
E supplied = (c + b · θ)T + f · I¯ + d · P (Wh) (2.38) f= (2.41)
η · Iann
where: where
E = the annual total energy use (Wh) sd = the shading factor of the external environment
T = the length of the heating season (h) (dimensionless)
P = the length of the non-heating season (h) Td = the transmission factor of the glazed part of the
c = a temperature-independent rate of energy windows (dimensionless)
consumption (Wh/h) during the heating season Ad = the glazed area of the window (m2 )
d = a temperature-independent rate of energy Id = the solar radiation on the window surface (Wh/m2 )
consumption (Wh/h) during the non-heating season Iann = the average global radiation on a horizontal
b = the gross heat-loss factor of the building unit (W/K) surface at the geographical location of the
f = the solar aperture of windows (m2 ) building (Wh)
θ = the mean sampled temperature difference
between internal and external environments (K) The gathering and processing of data
I¯ = the average sampled solar radiation (Wh/m2 ) Given the solar radiation I, the length of the heating
and non-heating seasons T and P, and the number of
The terms c, b, f and d are called energy parameters and have degree-hours θ · T from climate files (from the nearest
specific values for each building. climate station), the energy parameters can be determined
by regressing energy consumption with outdoor climate.
Physical interpretation of the energy parameters Energy consumption is obtained from the bills. By means
of interpolation, billed values that are obtained during
The ES method is in essence a black-box approach, giving
different time periods are synchronized.
limited information on how energy is consumed. However,
The part of Equation 2.38 that is climate dependent
vital information can be obtained if the energy parameters
involves the first two terms on the right-hand side of the
are interpreted.
expression. By setting the d · P term aside and dividing the
The gross heat-loss factor of the building unit is
remainder of the expression by the length of the heating

n
season, T, a new formulation is obtained which gives the
(U i · Ai + ψi · i ) + 0.33(1 − ηvent ) · n · V
i =1 average energy use per unit time w (kWh/h) during the
b= (W/K)
η heating or the cooling season. An error term ε is also added
(2.39) such that
where W = c + b · θ + f · s + ε (2.42)
n
i=1 (Ui · Ai + ψi · i ) = is the heat-loss factor of the This is a static model, which is valid provided that the
building envelope, including effect of thermal
observation period is long enough for the heat stored in or
bridges (W/K)
released from the building to be negligible compared to the
n = the average rate of air changes (1/h)
total energy supplied during the period. With the observed
V = the enclosed volume of the building unit (m3 )
data on energy consumption W, the temperature difference
η = the efficiency of the heating/cooling systems
θ and the solar irradiation s, the parameters c, b and f are
(dimensionless)
estimated with linear regression. If the error value ε exceeds
ηvent = the efficiency of air-handling systems
what is acceptable, more complicated models may have to be
(dimensionless)
applied; see below.
The winter factor c is climate-independent and primarily Application of the static model implies that the observed
related to the number of occupants and their behaviour: data are based on average or cumulated measured values
that correspond to those of Table 2.14.
Ehot water − Epersons − Ehousehold Where permanent gauges exist and continuous
c= (2.40)
η·T communication and data transfer are possible, hourly
52 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 2.14 Observed values for the model is then obtained as

Energy signature Average values of Monitoring    
β (XT X)−1 XT W
method θ and s periods = (2.48)
γ zT (XT X)−1 XT W
Cumulated values of w
Static model Weekly, monthly At least 10 observations where β is the original OLS-coefficient vector and γ is the
predicted energy use per hour for z, a normalized year.
An expression for the variance of the prediction is
average and cumulated measured values can be used as
obtained by noting that the total variance for the expanded
input to the ES method. This requires a dynamic model and
model (Equation 2.48), σtot2 , is
introduces time t.  T  
The energy use W(t) during hour t is assumed to be a linear ε ε
function of variables at times t, (t − 1) and (t − 2). A simple 0 0 εT ε
dynamic ARX (AutoRegressive model with eXogeneous σtot2 = = = σ2 (2.49)
k + 1 − (p + 1) k −p
inputs) model is expressed as
which equals the total variance in the unexpanded model
W(t) + a · W(t − 1) = c 1 + b 1 · θ in (t) (Equation 2.45). Thus the estimated standard error (se) of
+ b 2 · θ in (t − 1) + b 3 · θ out (t) + b 4 · θ out (t − 1) the estimate g is

+ f 1 · s(t) + f 2 · s(t − 1) + ε(t) (2.43) se(g) = σ̂(1 + zT (XT X)−1 z)0.5 (2.50)

With steady-state conditions, the thermal parameters are with which a confidence interval is obtained as
estimated such that
g ± t(k − p − 1) α se(g) (2.51)
2
c = c 1 /(1 + a)
The width of the confidence interval (Equation 2.51)
b = (b 1 + b 2 )/(1 + a)
depends on the number of observations and on the variability
f = (f 1 + f 2 )/(1 + a) of the data during the observation period. To increase
the accuracy, the measurement periods should be chosen
The parameters are estimated by minimizing the sum of
in such a way that the temperature differences and the
squares of the error term ε(t).
solar irradiation vary as much as possible and the average
temperature difference and the average solar irradiation are
Equations for estimating the normalized energy consumption as equal as possible to those of the normalized year.
This section is directly taken from Westergren et al.8 and
describes the procedure for estimating the normalized THE AUDITING PHASE
energy consumption.
Assume that data are available for k time periods (i = The auditing phase is primarily a complementary tool to the
1, 2, . . . , k), for which average measurements have been processing of the collected billed information. It serves to
obtained, and consider the following regression model with gather descriptive information on residential buildings and
p explanatory variables and a constant term: building performance and to reveal the living patterns and
behaviour of the occupants.
W i = c + b · θ i + f · s i + · · · + ε i (2.44)

or in matrix form Development of the audit form

W = X·β+ ε (2.45) The audit form is adapted from the ELIB project.4 The
ELIB project involved inspection of 1,200 residential units
To estimate the total heating energy use during a normalized
across Sweden, which was performed by 18 auditors during
year, a method suggested by Salkever9 is used, whereby
a planned time span of 6 months, with an extension of
the least-squares estimates are obtained together with the
6 weeks. The task was intended, as well as determining
predictions for a normalized year.
the indoor climate of the Swedish housing stock, to assess
Let zT = [1 θ¯ s¯ . . .] be a vector of the averages of the information on the technical and health aspects and the
p explanatory variables for a normalized year. Then expand thermal performance of this stock. It also dealt with assessing
the regression models (Equation 2.45): the potential of making the buildings more energy efficient
       and less dependent on electricity. The ELIB protocol has
W X 0 β ε
= + (2.46) therefore been extensively tested and there are statistics on
0 zT −1 γ 0
the frequency of answers to each question.
By using the partitioned inverse formula for matrixes, the The ELIB auditors took two days to learn the protocol and
Ordinary Least Squares (OLS) estimate of the parameter how gauges were to be placed. A test run of 90 houses was
vector used to ‘get the protocol right’ and to ‘calibrate’ the staff.
  Within the first six weeks of the project, the auditors had
β
(2.47) to produce an audit report using a proper and acceptable
γ
protocol if they were to continue the task.
 Experimental methods for the energy characterization of buildings 53

An ELIB audit took around eight hours for multi-family • The investigation routine should be the same for the
buildings and four hours for single-family buildings. The whole survey.
inspection comprised: • Train the personnel who are to carry out the field studies,
to ensure uniformity of terminology and usage.
• a review of the plans of the building • Carry out a pilot survey, to test the inspection report
• inspection at the site forms and routines.
• filling in the audit protocol • ‘Calibrate’ the auditors. This is very important in the case
• setting up the measurement equipment. of visual inspections with few instrumental readings and
many subjective assessments.
• The inspections should be carried out within a short
This was for an ‘average’ inspection unit, where plans
period of time and by a small number of inspectors.
were available, that is in 70% of the cases.10 The cost
was around 4,500 SEK in the early 1990s (Tolstoy, 2000,
private communications). One of the largest problems Auditing steps
encountered was getting in contact with representatives
• The auditor agrees a time to contact the customers
of a household, agreeing upon suitable day/days for the
(occupants/owner) and together with them inspects the
audit and measurements of buildings that lacked plans. This
building.
took more time than was taken into account during the
• The auditor informs the occupants/owners about the
planning stages of the project. The eventual cost was higher
BEP and/or MEP procedures. This includes information
because of this time ‘inflation’.
on the requirements, the limitations, the various services
Another audit form that was reviewed was the Residential
and the inconveniences that may arise during the
Energy Consumption Surveys (RECS) form. RECS ‘provides
monitoring phase.
information on the use of energy in residential housing
• The work of the audit starts with investigation of plans
units in the United States. This information includes
and technical documents. This information may be
the physical characteristics of the housing units, the
available from the local authority or from the building
appliances utilized including space heating and cooling
owner. Copies of this information are required. The
equipment, demographic characteristics of the household,
information is filled in on the audit form. If the building
the types of fuels used, and other information that
is composed of parts that differ significantly in terms of
relates to energy use. The RECS also provides energy
building year or building technology, or if common
consumption and expenditures data for natural gas,
surfaces (internal walls) have the same insulation level as
electricity, fuel oil, liquefied petroleum gas (LPG), and
envelope components, the various parts can be
kerosene.’11
considered to be individual buildings. Semi-detached
RECS 1997 involved interviews and collection of data
houses are considered to be detached buildings.
for some 5,000 residences. 220 interviewers were trained
• Together with the building owner (or representative),
for three days. Most interviews were conducted over
an audit is performed on site. The representative is
the telephone. The questionnaire took on average 29
preferably a person with knowledge of heating, cooling,
minutes to fill in and the span of 15–45 minutes covered
ventilation, building technology and electricity. If the
85% of the interviews. An electronic version of the
representative is not appropriate for the task, a more
questionnaire allowed immediate default value calculations.
suitable person may be asked for (this may be tested by
RECS 1997 is fully documented by the Energy Information
means of simple questions on, for example, the heating
Administration.11
system). The person is required to have keys to fan
The audit form that was developed and proposed in this
rooms, laundries etc.
book has been designed to do the following:
• The audit should be conducted so as to check that the
plans and interviews agree.
• take into consideration that the auditor is on site • If the representative cannot give information during the
• bring out the living patterns and behaviour of the audit, this information should be provided as soon as
occupants possible after this, for example in a telephone
• gather billed and non-billed fuel quantities conversation.
• collect technical information on the building and the • During the audit, if there is sufficient illumination, a
energy conversion systems colour photograph of the exterior should be taken.
• be used for both BEP and MEP • In residences (apartments) subjected to MEP, an
• take around an hour to complete. occupant must always be present because the gauges will
be placed and precautions described. If the occupants are
Recommendations for audits to read the gauges, the auditor must provide instructions.

Important points to bear in mind during field surveys:12

The audit form
• Do not rely on reported data and drawings before the The protocol is a type of guide for collecting information
auditor has seen the building with his/her own eyes. (a copy is included as Appendix 1). In the future, it may
54 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

be convenient to write a computer program that stores and A9 The distribution of apartments in a multi-family
processes the information. A benefit of doing this would be building. The main reason for this is that a
that the rating outcome (and certificate) could be produced multi-family building can participate in the rating
on the occasion of the audit (BEP) or when the monitoring procedure as a whole. This is the case when an
campaign is terminated (MEP). apartment is to be rated, but there is no
Some points may have to be modified to suit country- sub-metering of the energy consumption, the
specific properties. For example, revision of the first page of energy being distributed from a central heating
the audit form will be necessary. Data lists and that context system in a way that does not allow for individual
in which each building and the site is identified will have to measurements.
be adapted to national nomenclature and definitions. The A10 The main ventilation system type. This is merely
administrative information requirements are as follows: informative, although within MEP it is useful for
describing the heat-loss factor of the residence.
• identification of the building, site and address
A11 Records from the most recent tuning of the main
• building owner, resident representative and maintenance
ventilation system, which, like the information in
operator (may be the same or different people)
A10, is only informative.
• date of audit
A12 For apartments in a multi-family building only: on
• auditor name and company.
which storey the apartment is situated. This gives
The present audit form was applied in the four test coun- qualitative information on outdoor climate
tries. After this the original form was extended to include exposure.
questions on glazed surfaces. A13 Qualitative information on outdoor climate
exposure for multi-family and varieties of
General building information multi-family buildings (see A5).
A14 Number of rooms and bathrooms. This
General building information is to some extent informa-
information may be needed for BEP default value
tive, but also contains data that may be required in the
calculations. The number of rooms excludes halls
rating methodology. These items of information are as
and garages.
follows:
A15 Tables for listing window features (see
A2 Building year. This parameter is important with Appendix 2). The number of glass panes
respect to national building codes. Primarily, the (i.e. single-glazed, double-glazed or triple-glazed
building year will give an indication of how well the window) is to be filled in along with an estimation of
building is insulated. the g factor and the curtain factor. The horizontal
A3 Year of major retrofit/conversion/extension. If the angle for each lateral section of each window is
building has been subjected to retrofitting actions included. These angles are used to predict
(such as additional insulation) or it has been the partial shading factors for external obstacles.
extended, assumptions on the basis of the building Finally, the total glazed area for the different
year will no longer be valid. Furthermore, the windows or a representative window is given.
thermal performance of thermal zones may be
different for the various spaces. Residents and behaviour
A4 Building location. The location of the building
has relevance to the climate data that should be Residents and their behaviour have a large influence on
applied and the influence of shading obstacles. energy use. The main point of this section is, therefore, to
A code for the closest climate station can be given. obtain data on the patterns that the residents have. Much
A5 The type of building that is being audited. of the information has to be obtained from the resident
A6 How many storeys there are in the building, and representative.
the number of storeys that are heated and cooled. B1 The number of people in the apartment and their
This applies to both single- and multi-family ages.
houses. For multi-family houses, this information is B2 The presence of residents in the apartment/single-
important if the building has central heating and family unit during the summer and the
cooling without sub-metering. winter seasons. This is to take into consideration
A7 How the building is utilized. This is important with vacations when nobody uses the building.
regard to energy consumption patterns. Energy use B3 The presence of residents during working days and
for non-residential purposes is to be excluded. weekends is estimated. This information is used for
A8 Lengths, areas and volumes. The most important metabolic heat calculations.
are the last two, the residential heated floor area B4 The representative is asked for an estimation of the
and the total volume, because the various energy indoor temperature during certain periods of the
consumptions are given as areal or volumetric day (winter season). The estimate should be made
entities. This information is taken from the plans if on the basis of the living space that is occupied
possible, but has to be verified by the auditor by most frequently, such as the living room.
means of a visual survey. Furthermore, if the heating system is shut off
 Experimental methods for the energy characterization of buildings 55

during certain periods, this should be noted. These C9 Whether the energy (electricity) supplied only to
details, though they may be erroneous, are the heat pump is sub-metered. If so, the meter
valuable to determine the magnitude of indoor number should be filled in.
temperature. C10 A general question to obtain information on the
B5 If an automatic setback temperature is applied main type of heat distribution.
(commonly for central heating units), this should C11 The last occasion on which the heating system was
be noted. It will influence the mean indoor tuned. This is to provide information for C12.
temperature. C12 Information on the temperatures and flows in the
B6 An estimation is made by the resident representative heat distribution system, which can be assessed
of the indoor temperature in the summer season. if appropriate information is provided under C11.
Again, as in B4, this applies to frequently occupied If there are meters present, these can be read.
rooms (living rooms). C13 The main types of heaters used (more than one
B7 If an automatic setback temperature for cooling is alternative).
applied, it should be noted. This will influence C14 Information on district heating.
the mean indoor temperature. C15 If tuning records are available, information
B8 The length of the heating and the summer seasons should be gathered from these documents on the
may be obtained directly from the audit. If not, primary efficiencies of systems.
default values will have to be used for the region,
as prescribed at a national level.
Cooling
Cooling often uses electricity, but can in a few cases use
Heating natural gas. District cooling is not common and is, for
C1 The main type of space-heating system is surveyed; example, in Nordic countries only used for offices. However,
more than one alternative is possible. it is included in the protocol as a future alternative.
C2 The rated power is established for the main types
D1 How the audited building is mainly cooled.
of space heating. Information on age (or estimated
Alternative 6 states no cooling. However,
age) should be available, primarily serving as a
alternative 4 indicates the use of fans (in ceilings)
basis for boiler efficiency default values. Also, the
without cooling. The use of fans is climate-
meter or energy supplier account (subscription)
dependent and their utility is considered to be
should be included so as to identify which fuel is
greater during summer than during the winter
used where. For solar heating units, the total
season.
collector area should be assessed.
D2 Rated powers of the various types are noted,
C3 If the systems have been tuned, information on
together with the age and the ‘fuel’. The power of
primary efficiencies ought to be collected. If not,
local units is summed. The age is an average value
default values from national codes must be used,
for unit ages.
for example on the basis of type and age.
D3 If a tuning protocol exists, the cooling efficiency
C4 Individual boilers may use different fuels. These
factor is noted, if possible. Manufacturer
must be assessed. A convenient way of finding out
information is also useful.
the preliminary distribution of the fuels is by
asking the representative for estimated values.
This may reveal values for fuels that are not billed. Domestic hot water
C5 Certain boilers using different fuels have automatic
The use of domestic water varies considerably between
change-over between types. Patterns may be
households. It is seldom that energy for heating domestic
revealed here.
water is sub-metered. For the purpose of determining this
C6 Supplementary heat sources are assessed. In some
energy, as in the case where the same fuel and boiler is used
cases, these are not billed and the auditor has to
for space heating, or with solar heating, it is advised that
find the energy consumption on the basis of
the partitioning of seasons is considered first. If this is not
estimates. Another problem is if, for example,
possible, water consumption data can be used to estimate the
an electrical cooker is used for heating
energy demand.
(for example, if the heating system is shut off
when nobody is home but both are turned on E1 Is there a central domestic hot-water boiler?
upon arrival). The cooker, which is normally a E2 Are meters for hot-water consumption available?
climate-independent energy consumer, is Usually, they are not.
now a climate-dependent heater. E3 The table is used to assess how hot-water heating is
C7 Does one or more heat pumps exist? This heat done for both the heating and cooling seasons.
pump may be a central unit or local units may be These may be different or there may be a
used for heating and cooling. combination of different types.
C8 The types of use of the heat pumps should E4 The rated power and age of any separate
be included. water boiler.
56 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

E5 The rated power and age of the central water boiler REFERENCES
(for multi-family building).
1. EN ISO 13790: 2004 Thermal Performance of Buildings
E6 The set-point temperature of the hot water is
– Calculation of Energy Use for Space Heating.
assessed, either by a reading of gauge or from what
2. Kotsaki E and Sourys G, 2000, Critical Review and State of the Art
is stated in manuals.
of the Existing Rating and Classification Techniques. Report, Group
E7 The size of the hot-water tank (measured or Building Environmental Studies, University of Athens, Greece.
estimated). 3. Vekemans G, Loncour X, Bradfer F and Crabbé C, 2000, National
Report of the SAVE Belas Project: Belgium – Final Report, Study for
the European Commission DG TREN Contract XVII/4.1031/
Appliances Z/99-261. Vito Center 2001/ETE/001-07, Belgium.
Most household appliances and lighting are electrical, 4. Norlén U and Andersson K (eds), 1993, The Indoor Climate in
and therefore the consumption is a part of an energy the Swedish Housing Stock. Document D10:1993, Swedish Council
bill. With few exceptions, freezers and refrigerators are for Building Research, Stockholm, Sweden.
5. ISO 9060:1990 Solar Energy – Specification and classifica-
electrical unless gas is used. The most common appliance
tion of instruments for measuring hemispherical solar
that might not be electrical is the cooker (hob and oven).
and direct solar radiation. ISO/TC 180/SC1. International
For this reason, only two questions are formulated in this
Organization for Standardization, Geneva, Switzerland.
section. 6. Vallenor U and Wikström L, 1984, Energiförbrukning i Byggnader
– Delrapport 4: Mätning av oljeförbrukning och temperature,
F1 What types of fuel do the hob and oven use. Meddelande M84:17, The National Swedish Institute for
F2 How many hot meals are on an average cooked Building Research, Gävle, Sweden (in Swedish).
each day at home? From this frequency, a default 7. Westergren K-E and Waller T, 1998, Virtual Housing Laboratory
value for the energy used for heating a meal is – A system for simulating the energy use for heating in single family
used. houses, Working Paper No. 1, University of Gävle, Sweden.
8. Westergren K-E, Högberg H and Norlén U, 1999, ‘Monitoring
energy consumption in single-family houses’, in Energy and
Buildings, 29, 247–257.
Electricity meter information 9. Sálkever DS, 1976, ‘The use of dummy variables to
Buildings may be equipped with more than one electrical compute prediction errors, and confidence intervals, Journal of
Econometrics, 4, pp 393–397.
meter. For this reason, it is important to assess which meter
10. Tolstoy N, Borgström M, Högberg H and Nilsson J, 1993,
is measuring the various electrical uses.
Bostadsbeståndets tekniska egenskaper – ELIB-rapport nr 6, Forskni-
ngsrapport TN:29, The National Swedish Institute for Building
G1 Assessment of the use of electricity in other
Research, Gävle, Sweden (in Swedish, available in English).
buildings if this is registered by the electricity meter.
11. Energy Information Administration, 1997, A Look at Resi-
G2 Electricity meter number or code. dential Energy Consumption in 1997, DOE/EIA-0632(97), Energy
G3 Are any of the meters measuring specific energy Information Administration, Office of Energy Markets and
use and which specific energy use is being measured End Use, US Department of Energy, Washington DC. See also
by each meter? http://www.eia.doe.gov/emeu/recs/contents.html.
G4 Are there external electrical devices that are not a 12. Tolstoy N, 1994, The Condition of Buildings – Investigation
part of the energy balance of the building? methodology and applications, TRITA-BYMA 1994:3, Doctoral
G5 A list of external electrical devices. Energy dissertation. Department of Building Materials, KTH,
consumption by these is estimated by finding the Stockholm, Sweden.
rated power of each device, multiplied by the run 13. Álvarez S and Gonzlez Falcon R, 2000, Thoughts on
time when these are utilized. Other estimation Rating Methodology, Escuela Superior de Ingenieros, DIE-
types are possible. Grupo Termotecnia, Universidad de Sevilla, Camino de los
descubrimientos s/n, Sevilla, Spain.
14. EN 304: 1992. Heating boilers – Test code for heating boilers
for atomizing oil burners.
Suppliers 15. Martin S, Wouters P and L’Heureux D, 1996, Evaluation of a
The energy and water suppliers are to be listed. The simplified method for the energy certification of non-occupied buildings.
Final Report, Save Help Project, Belgian Building Research
information required is the names, addresses and phone
Institute (BBRI), Brussels.
numbers of the suppliers. The supplier account (or
subscription) number of the customer (resident), the meter
number, the fuel type and the unit of the quantity delivered Further reading
must also be listed. The information on whether or not the
Norlén U, 1982, Temperaturundersökningen 1982 – Huvudinstruktion,
billed quantity is estimated or exact for the billed periods
PM 1982-01-29. Statens institut för byggnadsforskning, Gävle,
must also be filled in. The conversion factor from the Sweden (in Swedish).
quantity unit to kWh should be noted. Wouters P and Loncour X, 2000, EURO-CLASS - Position paper of
In the third table, the delivery dates and quantities the BBRI about the objectives of the project. Division of Building
according to bills from the supplier (number according to Physics and Indoor Climate, Belgian Building Research Institute
the first table) are recorded. (BBRI).
 57

CHAPTER 3
Energy normalization techniques
JAN AKANDER, SERVANDO ALVAREZ AND GUÐNI JÓHANNESSON

INTRODUCTION the end-user behaviour and will allow comparison between

different occupants in the same residence.
Normalization of energy consumption can be done for
In accordance with the definitions of Chapter 2, the
several reasons and by different means. For the classification
energy concepts described below are used throughout the
of buildings, the choice has been made to normalize energy
chapter.
use from three possible points of view. The objective of
the three normalization types takes into consideration what
input is available (specific energy use and temperatures) and Space-heating energy
for the reason that normalization is done.
Space-heating energy is the sum of the specific energy for
Normalization takes into account:
space heating on an annual basis.
• the size of the building
• the external environment climate DELIVERED SPACE HEATING
• the internal environment climate.
The delivered space heating energy is the sum of heat
delivered
The size of the building is taken into consideration so as to dissipated from heating units Q spaceheating .
create some variable that will allow comparisons of building
energy performance. Within this task, the choice was to use SUPPLIED SPACE HEATING
the heated floor area as the benchmarking criterion. The
Dividing delivered energy by the efficiency of the discharg-
use of heated floor area will relate energy use to the parts of
ing unit gives the sum of supplied space-heating energy
the residence that are utilized and conditioned. supplied
Q spaceheating .
Within the framework of the methodology, it is advisable
that the normalization with regard to building size is uniform
for all countries. However, national customs already offer a Space-cooling energy
diversity of normalization parameter types. The framework
Space-cooling energy is the sum of the specific energy for
has, therefore, been left open for this parameter to be
space cooling on an annual basis.
defined at a national level. The parameter, now the heated
floor area, can be changed to the protected volume, a
characteristic length etc. The use of national parameters, DELIVERED SPACE COOLING
and thereby national energy use values, allows comparisons The delivered space cooling energy is the sum of heat
within a single country. Use of size conversion factors will delivered
absorbed by cooling units Q spacecooling .
allow comparison across national borders. For example, the
conversion factor between the heated floor area and the
SUPPLIED SPACE COOLING
heated volume is building height. The present parameter,
heated floor area, illustrates how such a parameter is defined Dividing delivered energy by the efficiency of the discharging
supplied
and utilized within the framework. unit gives the sum of supplied space-cooling energyQ spacecooling .
Normalization with respect to the external environment
climate takes into consideration annual variations. By
placing the building within a reference, or normal, climate
Global energy
for the site, not only will the performance of the building Annual global energy is the sum of specific energies.
be comparable with the performance at other times, but How external energy should be handled deserves special
it will also be possible to compare the performance with attention.
that of other buildings subject to the same external climate.
For a building that has undergone only normalization with
DELIVERED ENERGY
respect to the external environment it is assumed that the
same indoor conditions prevail. Delivered global energy is the sum of delivered specific
Normalization of the internal environment climate can energy and is straightforward in this sense. However, the
delivered
be done in terms of a set of predefined indoor conditions. delivered external energy Q external must be handled correctly.
This type of normalization will, in fact, standardize a part of If this energy is delivered outside the considered building
58 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

delivered
envelope, Q external = 0 since by definition, this is not delivered Complementary information that may be obtained is as
into the building space and will not affect the heat balance of follows:
delivered
the building. In the case where Q external is delivered within the
building envelope, but outside the considered apartment, it • the heated floor area assessed from the audit and the
will be included in delivered global energy. plans (or other necessary data to determine the size
As an example, consider an apartment that is to be rated parameter)
in a multi-family building. The apartment is billed for the • indoor temperatures obtained from the occupants or
energy use within the apartment, but it will also be billed an from representatives of the maintenance operator
extra fee for shared costs for the heating of corridors and • set-point temperature of the main energy conversion
staircases and for elevator operation. This extra fee corre- systems and energy distribution devices, if available
delivered
sponds to Q external , which is delivered within the envelope • external environment climate for the period covered by
of the multi-family building and which will be included in bills; this may be obtained from the closest climate station
the delivered global energy. However, if an outdoor pool is • reference external environment climate, which is
regularly heated, the external energy delivered to the pool determined at a national level.
delivered
will be considered to be zero Q external = 0.
Normalization of BEP output values is done on the basis
SUPPLIED ENERGY of the heated floor area and the external climate only.
Normalization is not done for the internal environment
Supplied energy is the sum of all supplied specific energies,
because the indoor temperatures are considered to be
and this also includes all energy supplied to external devices.
unknown. End-user information may be erratic, and the
situation may be such that set-point temperatures of
NORMALIZATION OF BEP AND MEP OUTPUT the heating and cooling units are not accessible.

The proposed methodology provides a normalization

platform that depends on which experimental protocol has MEP
been employed. Primarily, it is the availability of data input
The Monitored Energy Protocol (MEP) offers an insight
that determines which types of normalization can be applied
into the internal environment temperature during the
and the principles of the rating scheme. Results from BEP
season that measurements were conducted, although the
and MEP are treated in the same way, so as to make the
monitoring period may be considered short from the view-
normalized results from different residences comparable.
point of the energy involved. As well as information from
Both are normalized with respect to the heated floor area
the measurements and the audit, information that can be
and the external climate.
assessed elsewhere is as follows:
The reason for excluding normalization of the internal
climate in the rating procedure is twofold:
• the heated floor area assessed from the audit and the
• BEP does not always offer reliable information on plans (or other necessary data to determine the size
internal environment conditions. parameter)
• The rating scheme is based on normalized energy use • indoor temperatures obtained from the occupants or
with respect to the building size (floor area) and from representatives of the maintenance operator
the external climate. The indoor conditions and and from the monitoring period
consequently energy use are dependent on occupancy • set-point temperature of the main energy conversion
behaviour. Low-energy consumers, who prefer to have system and the energy distribution devices, if available
a low indoor temperature, should not be ‘punished’ in • external environment climate for the period covered
the rating process. In contrast, high-energy consumers by bills; this may be obtained from the closest climate
should receive a worse rank. The rating method station – monitored external environment variables
evaluates energy use of the building including the can be compared with those from climate stations,
effect of occupancy behaviour. so that systematic deviations can be detected
• reference external environment climate, which is
Because the indoor temperature is assessed in MEP, determined at a national level.
normalization of internal climate can be performed together
with the other normalization types. The outcome of this Normalization of MEP output values is done on the basis
normalization will not be rated and should thus inform of the heated floor area and the external climate, in the
the occupants how their behaviour compares with that of same manner as for BEP outputs. This makes normalized
‘standard occupants’. results from the two protocols comparable. Moreover, with
the insight into internal climate temperature, normalization
of internal climate can be performed as a service to the
BEP occupants. This gives the occupants information on their
The Billed Energy Protocol (BEP) makes use of bill energy use compared to that of ‘standard occupants’ or to
information to serve as input to the rating procedure. ‘standard indoor conditions’.
 Energy normalization techniques 59

NORMALIZATION WITH RESPECT TO THE θe = mean outdoor temperature on a daily basis [◦ C]

EXTERNAL ENVIRONMENT CLIMATE tstart = start day of the season
tend = end day of the season.
Normalization with respect to the external environment
climate is performed for both BEP and MEP. This The following remarks should be noted:
normalization is preparatory to the rating procedure: the
residence is adjusted for the reference climate, which • The numeric values of θHDD_base , θCDD_base and the
will bring changes to space-heating and space-cooling definition of the length of the heating/cooling season
energy. The other specific energies may also receive normal are left to be determined at a national level.
values, for example default values for a ‘standard’ or • The degree-days are calculated on a daily basis.
‘normal’ energy use for a normal occupancy. This type of • The summing is done for non-negative differences,
normalization cannot be done for the output from BEP, which indicate the presence of a heating or cooling
the primary reason being that the temperatures of the requirement.
internal environment are not known. Normalized values • Such calculations are made for the year with the actual
of this type influence the internal environment and can climate (HDD and CDD), and for the year with the
only be done within MEP, when monitoring of tempera- reference conditions (HDD N and CDD N ).
tures has been performed. Therefore, the internal climate is
not normalized prior to the rating procedure. The internal The most straightforward method of normalization based
climate conditions during the prevailing year are assumed on the degree-day concept is to use the ratio between the
to be the same as those for the reference year. Implicitly, reference and the prevailing degree-days. The following
climate-independent variables are assumed to be the same expressions are used:
for the two time periods under consideration. N delivered delivered HDD N
Three normalization procedures are proposed, and the Q spaceheating = Q spaceheating · (3.3)
HDD
choice of normalization method (or methods) to be used
will be determined at a national level. These three are: N delivered delivered CDD N
Q spacecooling = Q spacecooling · (3.4)
CDD
• the heating and cooling degree-day method (DDM)
• the modified utilization factor method (MUFM) LIMITATIONS OF THE MODEL
• the Climate Severity Index (CSI).
The degree-day concept has several limitations:

The heating and cooling degree-day method • First of all, only temperature is taken into consideration.
For instance, there is no solar radiation taken into
The heating and cooling degree-day method is a traditional
account.
method that has been in use for decades, in both the aca-
• There is also a problem with defining the magnitude of
demic and the industrial worlds. It is a well-known method
the base indoor temperature and the length of each
and is easy to use, although there are several variants.
season for individual residences.
In general, the concept primarily builds on the
• The use of degree-days is considered to be adequate if
temperature difference between a base indoor temperature
internal gains and solar gains do not significantly
and the outdoor temperature, multiplied by the duration
influence the heat balance of the building. This is only
of the temperature difference. It is quite common for the
the case for Nordic countries, when the space heating
length of heating and cooling season to be pre-determined.
energy forms a large part of the whole energy use.
The base indoor temperature is also prescribed, with
different values and definitions in various countries.
The cooling degree-day method has severe limitations in
Expressed in equation form, the number of degree-days
that solar radiation is not accounted for and that cooling is a
calculated on a daily basis, are:
non-linear phenomenon. The degree-day concept assumes
that the heat-loss factor is constant and that heat loss is
• for the heating season
proportional to the temperature difference. In most cases,
tend
HDD = (θHDD_base − θe ) (3.1) cooling is only applied in a few rooms and for a limited
tstart
period of time, in general during the peak hours during
• for the cooling season the day.
tend
CDD = (θe − θCDD_base ) (3.2)
tstart TOLERANCES OF THE MODEL

The normalized values that are obtained should be tagged

Here, the symbols denote:
with an error band. For the degree-days, three calculations
HDD = degree-days for heating [◦ C · days] must be performed. These are, for the example of heating,
CDD = degree-days for cooling [◦ C · days] degree-day calculations for the base temperature θHDD_base
θHDD_base , θCDD_base = base indoor temperatures for and for θHDD_base ± 3◦ C. This applies to both the prevailing
the considered period [◦ C] and the reference climates. The results of normalization can
60 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

be given within the bandwidth of: the part of the internal heat gains that is useful in obtaining
N delivered N delivered N delivered the set-point temperature within the space. The rest of the
min Q spaceheating < Q spaceheating (θHDD_base ) < max Q spaceheating
heat gains are considered to create an internal temperature
(3.5)
that is above the set-point temperature.
where A re-formulation, in which the space temperature is
N delivered N delivered now the set-point temperature (spt) rather than the actual
min Q spaceheating = min[Q spaceheating (θHDD_base + 3), temperature of the space, can be made such that
N delivered
Q spaceheating (θHDD_base − 3)] at_spt delivered delivered
Q heatlosses = Q spaceheating + ηUF (Q appliances + Q metabolic + Q solarheat )
N delivered N delivered (3.7)
max Q spaceheating = max[Q spaceheating (θHDD_base + 3),
The utilization factor ηUF is determined either on a seasonal
N delivered
Q spaceheating (θHDD_base − 3)] or a monthly basis, depending on the periodicity of bills
that have been collected.
Analogously, the same types of equations apply to space at_spt
Note the differences in energy losses, Q heatlosses and Q heatlosses .
cooling.
These are related to the temperatures that are used in the
calculations, where
The modified utilization factor (MUF) method at_spt
Q heat losses = H(θspt − θe ) · t (3.8)
The modified heating degree-day method was proposed for
two reasons: Q heat losses = H(θi − θe ) · t (3.9)
with H denoting the heat loss factor of the building, ts the
• The normalization method has to take into consideration period of time, θe the mean external temperature, θspt the
the solar gains. set-point temperature of the heating system and θi the mean
• The set-point temperature of the main heating system indoor temperature.
has to be used since this is a temperature that can be The procedure of prEN ISO 13790 calculates heat losses
assessed from an audit, without the need of detailed relative to the set-point temperature. The procedure makes
measurements. use of the so-called gain–loss ratio γ during a considered
period of time, defined such that
The space-heating requirement, as specified earlier, is
delivered
dependent on four components, heat delivered into the Q appliances + Q metabolic + Q solarheat
γ= at_stp
(3.10)
space by heating systems, appliances and lighting, solar and Q heat losses
metabolic heat. On the basis of a steady-state approach, the
heat losses by means of transmission and ventilation will The utilization factor is calculated using the gain–loss ratio
equal the heat delivered into the space, such that so that
delivered delivered 1 − γa
Q heat losses = Q spaceheating + Q appliances + Q metabolic + Q solarheat ηUF = if γ = 1 (3.11)
1 − γ a+1
(3.6)
a
ηUF = if γ=1 (3.12)
The indoor temperature that is obtained over time is a a+1
result of when, where and how much energy is delivered where
into the space, the heat-loss factor and the heat capacity of   
required
the residence. The indoor temperatures of thermal zones in Q heatloss = Q spaceheating + ηUF Q appliances + Q solarheat
the building are usually unknown during the season under  
consideration. The most reliable temperature data that can + Q metabolic for month-wise calculations
be assessed during the heating season, without monitoring, is τ
the set-point temperature of the heating system. Therefore, a = 0.8 + for seasonal calculations
28
calculations that serve to determine delivered space-heating The time constant τ is defined as the ratio of the effective
energy, both actual and normalized, will make use of the heat capacity to the heat-loss factor of the building.
set-point temperatures. A utilization factor for internal and
solar heat gains is introduced. By means of the utilization NORMALIZATION OF SPACE HEATING ENERGY
factor and adjusting the indoor temperature to be the
set-point temperature, normalization of the space-heating Normalization of the space heating energy can be performed
energy is possible. with the utilization factor concept as presented above. Some
modifications have to be made, since a ‘backward’ prEN
ISO13790 procedure is applied. The steps of this procedure
INTRODUCTION OF THE UTILIZATION FACTOR FOR HEAT GAINS
are illustrated below.
If the indoor temperature is not known, the set-point prEN ISO13790 is based on a descriptive model, which
temperature of the heating system can be used in the assumes that characteristics and data on the building, the
calculations. The calculation has its roots in prEN ISO energy systems, end-user behaviour and occupancy are
13790,1 which makes use of the so-called utilization factor known or prescribed, together with the reference external
for internal heat gains. The utilization factor is a measure of climate. From the viewpoint of this methodology framework,
 Energy normalization techniques 61

Table 3.1 Time constants (hours) for various building types with the following equations for iteration step i:
consideration of the internal mass. The table may be extended by 1
γ i +1 = Qheat losses i
(3.14)
evaluating the amount of insulation. The values in this table are only for − (1 − ηUF )
delivered
Qappliances +Qsolarheat +Qmetabolic
illustrative purposes
Building type Light-weight Medium-weight Heavy-weight Then,
Single-family 20 30 60
i +1 1 − (γ i +1 )a
house ηUF = if (γ i +1 ) = 1 (3.15)
Terrace house 25 40 80
1 − (γ i +1 )a +1
Apartment 30 50 90 i +1 a
ηUF = if (γ i +1 ) = 1 (3.16)
a+1

it is the output of this descriptive model that has been The computational loop will give the final value for η∗UF .
assessed in terms of billed quantities, for the actual climate.
Determining the potential heat losses at
the set-point temperature
Background information from the audit
The final value of η∗UF will now be used to estimate what the
The auditor chooses the time constant of the residence
heat losses would be if the internal temperature were equal
at the time of audit. Table 3.1 shows the alternatives, and
to the set-point temperature. Expressed in equation form,
the magnitudes of the time constants are to be taken from
this corresponds to
representative buildings in a particular country.
Q heat losses = Q spaceheating + η∗UF (Q appliances + Q metabolic + Q solarheat )
at_spt delivered delivered
From the audit, information is gathered for the solar
apertures and the orientations. Based on national or regional (3.17)
areas, there must be computational codes for performing
solar radiation calculations. It may be convenient to use a
development of national standard codes for this purpose, or
Heat losses for the reference year
to adapt routines from building simulation programs. The next step is to normalize heat losses with respect to the
external climate. The base indoor temperature is in this case
the set-point temperature. This is more or less a traditional
Estimation of heat losses at the prevailing
temperature degree-day calculation for month-wise or seasonal average
temperatures, as follows
At the prevailing indoor temperature, an estimation of
N at_spt at_spt
(θspt − θeN )
the heat losses by transmission and ventilation is made. Q heat losses = Q heat losses · (3.18)
The estimated right-hand side variables for a month-wise or (θspt − θe )
seasonal period will give
delivered delivered
Normalization of space-heating energy
Q heat losses = Q spaceheating + Q appliances + Q metabolic + Q solarheat
(3.13) From this step on, the calculations are the traditional prEN
ISO 13790 calculations. However, preparatory calculations
where are performed to estimate the solar energy for the reference
delivered
Q spaceheating is the estimated sum of the delivered space climate that is delivered into the building (and, if required,
heating energy from conversion systems normalized values from appliances and metabolic heat). The
(from BEP and MEP) gain–loss ratio for the normalized conditions is now
delivered delivered N
Q appliances is the estimated sum of the energy delivered Q appliances + Q metabolic + Q solarheat
from appliances and conversion systems γ =
N
N at_stp
(3.19)
Q heat losses
(from BEP and MEP)
Q metabolic is the estimated energy delivered from the The normalized utilization factor is calculated using the
occupants calculated from the dissipated heat normalized gain–loss ratio, here expressed as
per occupant and the time of presence
N 1 − (γ N )a
Q solarheat is the estimated solar energy delivered into ηUF = if (γ N ) = 1 (3.20)
1 − (γ N )a +1
the building, based on climate data
N a
calculations (national methods) and solar ηUF = if (γ N ) = 1 (3.21)
aperture estimations (from BEP and MEP). a+1
The normalized space-heating energy is calculated from the
set of variables as follows
Calculation of the modified gain–loss ratio
N delivered N at_spt N delivered N
Q spaceheating = Q heat losses − ηUF (Q appliances + Q metabolic + Q solarheat )
The modified gain–loss ratio is now calculated for the
(3.22)
period considered. The calculation is iterative, assuming an
i =1
initial value for ηUF . This value can be set to equal unity and This is the normalized delivered space-heating energy.
five iterations are enough. The iteration will loop through The supplied space-heating energy for each fuel type is
62 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

the part that it has contributed to the specific energy for TOLERANCES OF THE MODEL
space heating so that
The tolerances of the model are not trivial to assess in this
N delivered
N supplied
Q spaceheating method. This is because many of the variables are linked
Q spaceheating = (3.23) to each other, as will be shown in the following text. The
ηsh
calculations for tolerances will have to be specified on a
NORMALIZATION OF SPACE-COOLING ENERGY national basis; since the influence of climate variables in this
model will be non-linear. Below, an example is given, but
As in the case of heating, the cooling season involves
the parameters will have to be changed to relevant national
an energy use for maintaining a desired internal climate.
values.
However, even in the most idealized cases the quantity
If the set-point temperature is assessed during the audit,
that is used is non-linear, which makes assessment and
the calculations of ‘degree-months’ can be performed such
normalization more complicated. Though the heat balance
that the base temperature corresponds to θspt and for
equation (see below) initially seems straightforward, deter-
θspt ± 1◦ C. If the set-point temperature is not known, the
mination of each variable becomes exceedingly difficult to
‘degree-month’ part of the calculations should be done for
master with low inaccuracy:
the base temperature ±3◦ C.
delivered delivered
Q spacecooling + Q heat losses = Q appliances + Q metabolic + Q solar (3.24) A sensitivity analysis must be done on the basis of the time
constant, with τ ± 25%.
Furthermore, space cooling is seldom applied to the whole The influence of solar radiation must be assessed where
residence. Use is made of cooling in rooms that are occupied, the solar apertures vary in size, As,n ± 50% or where
and usually only for a limited time during the occupants’ Q solarheat ± 50%, in combination with the base tempera-
presence. For these reasons, space cooling will strictly be ture deviations.
limited to analysis of bills (supplied energy) and a different The results of normalization can be given within the
normalization scheme will be applied to that used in the case bandwidth of
of space heating.
N delivered N delivered N delivered
At the present time, there is no European norm that min Q spaceheating < Q spaceheating < max Q spaceheating (3.25)
covers the space cooling energy requirement in the same
way as prEN ISO 13790 does for space heating. Such where
procedures are under development and similarly make use N delivered N delivered
min Q spaceheating = min[Q spaceheating (θspt ± 3,
of the utilization factor, for example the Dutch building
code NEN 2916 (for commercial buildings).2 Q solarheat ± 50%, τ ± 25%)]
N delivered N delivered
max Q spaceheating = max[Q spaceheating (θspt ± 3,
LIMITATIONS OF THE MODEL

The modified utilization factor method has several Q solarheat ± 50%, τ ± 25%)]
limitations. First, a central set-point temperature may not Analogously, the same types of equations would be used in
be available in the building considered. If so, national space-cooling applications.
default values will have to be applied, preferably taking
into consideration information from the occupants or by
A pre-study
measuring temperature profiles during the day.
Another limitation is the estimation of solar radiation The influence of each variable is not trivial, since these
that is delivered through apertures. Shading factors, curtain influence one another. A pre-study is given in this section,
factors and user behaviour have great impact on solar showing how the variables affect the outcome of the
irradiation. Not only do these vary over the seasons, but procedure, which is the space-heating energy. In order to
they also vary considerably throughout each day. However, facilitate the presentation, some definitions are given below.
within the frame of normalization calculations, the effect of The relative change in space heating is based on the space-
an erroneous solar term will only affect the normalization of heating energy that is assessed, whereas the normalized
the heat loss; the same error will afterwards be subtracted space-heating energy is the outcome of the MUF method.
from the final results (see Equations 3.17 and 3.22). The relative change that the normalized space heating gives
Finally, the use of a utilization factor can be doubtful. In rise to in comparison to the actual space-heating energy is
this application, the time constant is unknown and will be formulated such that
estimated on the basis of the auditor’s intuition (calculations N delivered delivered
based on plans would be quite cumbersome). Moreover, Q space heating − Q space heating
change = delivered
(3.26)
the underlying theory of the utilization factor is based Q space heating
on generalization of data from numerous buildings, which
makes results for one specific building uncertain. On the After some mathematical operations, the following
other hand, based on the limitations within the experimental expression can be obtained:
protocols BEP and MEP, the order of the deviations that may N
arise from this technique will probably not be greater than β − ηUF · ρ · γ
change = (3.27)
the uncertainties within the specific energies. 1 − ηUF · γ
 Energy normalization techniques 63

where
N
θ spt − θe
β = the ratio between temperature Ndifferences θspt − θe

Change in space heating (%)

Q 200
ρ = the ratio between heat gains, Q gains
gains Q gains 150
γ = the gain–loss ratio of the actual period, at_spt
Qlosses
100
N 50
and ηUF (γ, τ) and ηUF (γ N,
τ) are dependent on the time
constant τ and the gain–loss factors of the actual and the 0

normal periods. The relationship for the gain–loss factors is -50

that γ N = β γ.
ρ
-100
3
The normalized space heating will be roughly dependent 1.5
2
on the gain–loss ratio for the actual period, the ratio of heat
1 1
gains, the temperature difference ratio for the two periods γ (dimensionless) β (dimensionless)
0 0.5
considered and the time constant. In a computational study,
these sets of equations were used to derive the order of Figure 3.1 The relative change in space heating as a function of γ and β
magnitude of the change in space-heating energy. when ρ is kept at unity. The time constant of the building is 30 hours
The first set of calculations was intended to show the
influence of the temperature of the external environment.
The ratio between heat gains ρ was set to unity. The results Change in space heating (%)
are shown in Figures 3.1, 3.2 and 3.3, where the time 300
constant has been given the value of 30, 50 and 70 hours
respectively. Calculations were based on the heating 200

requirement for the heating season (not on a month-wise

150
basis).
Figure 3.1 shows the relative change in space-heating 0
energy as a function of the gain–loss ratio for the actual year
and the variation of the temperature difference ratio. For -100
3
the case where the normal year is the same as the actual 1.5
2
year, β is unity and the relative change is zero for all γ. (By
1 1
dividing γ by β, β can be substituted by γN on the x axis.) γ (dimensionless) β (dimensionless)
For a colder normal year, β increases and will lead to a larger 0 0.5

space-heating requirement than for the actual year, since Figure 3.2 The relative change in space heating as a function of γ and β
this means a reduction in γN . If the γ value is followed with when ρ is kept at unity. The time constant of the building is 50 hours
changing β, the corresponding change in space heating can
be observed.
A case in Sweden is taken as an example; the gain–loss
ratio for the heating season is of the magnitude of 0.7.
Change in space heating (%)

400
This involves some ±70% if the temperature difference
ratio varies ±50%. The temperature difference does not 300
vary this much from one heating season to the other.
200
In Figure 3.2, the results for a building with the time
constant 50 hours are presented. The relative influence 100

of changes in the external temperature is larger than for 0

the ‘thermally lighter building’. This is because the time
-100
constant is a function of the thermal mass of the building and 3
also of the level of insulation. A large time constant means 2 1.5

more insulation and thus less absolute energy use when β is 1 1

unity. The relative change in space heating will therefore γ (dimensionless) β (dimensionless)
0 0.5
increase for larger time constants, although the absolute Figure 3.3 The relative change in space heating as a function of γ and β
change is less than for the cases with a smaller time constant. when ρ is kept at unity. The time constant of the building is 70 hours
The largest relative changes are found for large time
constants and when the temperature of the outdoor
environment has dropped and heat gains are large during gains, fairly constant. The ratio will reflect changes in
the heating seasons. This effect is clearly seen in Figure 3.3, solar gains, together with any climate-independent internal
where the building time constant is 70 hours. gains.
The same type of analysis has been done for the case Again, the greatest relative changes in space heating are
where the ratio of heat gains ρ is varied and the temperature obtained with large values of γ in buildings with a large
difference ratio is kept at unity (Figures 3.4, 3.5 and 3.6). time constant. The largest impact of change in heat gains
In practice, the heat gains are, with the exception of solar is found when the heat gains are decreased.
 64 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

more effectively, and when heat gains are reduced and

limited, this will have to be compensated by space-heating
energy.
Change in space heating (%)

150

100
The Climate Severity Index
50 In terms of the heating requirements of buildings, two
climatic conditions can be considered to be ‘identical’ when
0 the heating energy consumption of a certain building is the
3
same under the two climatic conditions.
-50
0.5 In this context, the concept climatic conditions must be
2
understood in a broad sense:
1 1
ρ (dimensionless) 0
γ (dimensionless
(dimensionless)
1.5
• two actual years at the same location
Figure 3.4 The relative change in space heating as a function of γ and ρ
• two actual years at different locations
when β is kept at unity. The time constant of the building is 30 hours
• an actual year and the reference year at a certain location
• two reference years at two locations respectively
• and so on.

These definitions can obviously be translated to the cooling

Change in space heating (%)

300
0
requirements and logically it can happen that two climatic
0
200 conditions can be ‘equal’ in terms of heating requirements
but ‘different’ in terms of cooling requirements. The
0
100 reciprocal situation (equal for cooling and different for
heating) is also true.
0
The idea can be extended by saying that, for a certain
3 building, a climatic condition is x times more severe than
-100
0
0.5
5 2
another when the energy consumption of this building is x
1
times greater under the former condition than under the
1
ρ (dimensionless) γ (dimensionless
(dimensionless) latter.
1.5
1 5 0
One of the ways to characterize the climatic dependency
Figure 3.5 The relative change in space heating as a function of γ and ρ of the heating or cooling requirements of buildings in the
when β is kept at unity. The time constant of the building is 50 hours framework of the Euroclass project,3 is the so-called Climatic
Severity Index (CSI), which makes it possible to compare
the ‘severity’ of different climatic conditions. The more
severe the climate, the bigger the energy requirements of
the buildings and, consequently, the bigger the value of
the CSI.
Change in space heating (%)

500
There are, in principle, two CSIs, one for the heating
400
season and other for the cooling season. The CSI as it is
300 used is conceptually a positive number. When the CSI is zero
200 or negative, it can be assumed that there are no significant
100 energy requirements (for heating or cooling) under these
0 climatic conditions.
-100 3
0.5 2 DETERMINATION OF THE CSI
1 1
ρ (dimensionless) 0
γ (dimensionles
(dimensionless) Let us suppose that the heating requirements are calculated
1.5
for a given building under different climatic conditions that
Figure 3.6 The relative change in space heating as a function of
correspond either to different locations or to different years
γ and ρ when β is kept at unity. The time constant of the building is
at a certain location or to a combination of both. The set
70 hours
of values obtained is divided by the value corresponding
to a, say, pivot or representative climatic condition (see
Figure 3.6 shows that the gain ratio has the largest impact Table 3.2). The reduced heating requirements for this
for buildings with a large time constant. A severe reduction climatic condition will obviously become unity (1) and the
in gain ratio leads to a large increase in space heating reduced values for the other climatic conditions will provide
compared to the case of small time constants. This is because the ratio between these and the heating requirements that
the building with a large time constant uses heat gains would be obtained for the reference climate.
 Energy normalization techniques 65

Table 3.2 Heating energy demands used in the calculation of the tational simulations. The results for every combination
winter CSI run for the selected climatic conditions are calculated and
Burgos −21.373 1.9 provide the CSI for each combination. The average of the
Soria −18.427 1.6 CSIs for all the combinations is the CSI for this sector.
Segovia −15.945 1.4 This resulting CSI is finally correlated with the con-
Madrid −11.412 1.0 ventional climatic variables corresponding to all the climatic
Bilbao −11.097 1.0 conditions selected (degree-days, monthly average global
Barcelona −7.782 0.7 solar radiation, insolation etc.) in order to make the concept
Seville −3.769 0.3 applicable to climatic conditions different from those used
Cádiz −2.362 0.2 to perform the simulations.

In this way, an index of the relative influence of the climate on

EXAMPLE AND ROBUSTNESS OF THE CSI
the heating consumption of a building is obtained. For instance,
a climatic condition that yields a value of 0.6 implies that Two CSIs (one for heating and one for cooling) were
the heating requirements of the building are 60% of those calculated in 1996 during the project for developing
of the pivot. A value of 1.4 means that the heating require- the Spanish Energy Labelling for Social Housing (CEV
ments are 140% bigger than in the pivot climatic condition. project hereinafter).4 These two indexes were obtained
The absolute climatic influence on the heating require- from simulations based on the ‘Passport +’ computer code.
ments of a building depends strongly on the building The CSI was applied to the capitals of the 50 Spanish
characteristics, but the relative climatic influence (as it has provinces and the climatic condition used to make the values
been defined above) is quite independent of factors such equivalent was the TRY of Madrid.
as the quality of the envelope, the window to wall ratio or Figure 3.7 shows the correlation of the CSIs for summer
the orientation of the building. The only significant factor and winter as a function of degree-days and the monthly
remaining is the use of the building, mainly because of the average solar radiation.
internal gains. The winter CSI correlation was obtained as follows:
With the assumption that buildings in the same sector (for
instance residential) have a similar value for their internal CSI = a ∗DD + b ∗n/N + c ∗DD 2 + d ∗n/N 2 + e (3.28)
gains, it is possible to calculate the CSI corresponding to
each sector for a particular country or a wider geographic where
area.
The preparatory steps are: DD = the degree-days for heating with a base temperature
of 20◦ C for the months of January, February and
1. selection of the pivot climatic condition December
2. selection of a sample of the building typologies n/N = the ratio between the actual insolation hours and
common in the geographic area the maximum insolation hours for that latitude for
3. selection of climatic conditions covering the different the months of January, February, and December
climates of the area – typically typical meteorological year a = 2.395E − 03
(TMY) or test reference year (TRY) of selected locations. b = −1.111E + 00
c = 1.885E − 06
Next, the energy requirements of the buildings under d = 7.026E − 01
different scenarios, for example the orientation of the main e = 5.709E − 02
façade, different UA values, etc., are calculated via compu- R 2 = 0.99

Figure 3.7 CSI correlation for winter and summer seasons

66 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

2.0 SOUTH
Normalised Climate for 15

1.8 2
1.6
Different Buildings

1.4
1.2 1.5
1.0
0.8
1
0.6
0.4
0.2 0.5
0.0
0.0 0.5 0.5 1.5 2.0
Climatic Severity Index (CSI) 0
Figure 3.8 Correlation between the CSI for the heating requirement of 0 0.5 1 1.5 2
15 representative Spanish residential buildings in 50 locations Figure 3.9 Correlation for a Belgian building in Spanish locations

The summer CSI correlation was obtained as follows: SOUTH

4
CSI = a ∗DD + b ∗n/N + c ∗DD 2 + d ∗ n/N 2 + e (3.29)

where 3
DD = the degree-days for heating with a base temperature
of 20◦ C for the months of June, July, August, and
September 2
n/N = the ratio between the actual insolation hours and
the maximum insolation hours for that latitude for 1
the months of June, July, August, and September
a = 1.090E − 02
b = 1.023E + 00 0
c = −1.638E − 05 0 1 2 3 4
d = −5.977E − 01 Figure 3.10 Correlation for a Belgian building in Spanish locations (filled
e = −3.370E − 01 symbols) and at the sites of Marseille, Paris, Lyon, Uccle and Stockholm
R 2 = 0.97 (open triangles)

In the Euroclass project, the applicability of these

expressions was checked for three effects: Figure 3.10 shows the effect of adding new climates
(Marseille, Paris, Lyon, Uccle and Stockholm) represented
• the influence of the computer code used with open triangles. The results are acceptable, except when
• the influence of the building type the correlation is applied in Stockholm with a climate
• the influence of the range of climatic conditions. that is clearly colder than that of Spain.
In summary, the correlation developed in the CEV
In all the cases, the CSI correlations developed for the project is consistent when applied in a different climatic
CEV project proved to be good enough to characterize and constructive context and could be used to cover the
the influence of the climatic conditions on the heating or majority of EU countries. However, it is obvious that a
cooling requirements. better approach would be obtained for CSI if, at a national
As an example, Figure 3.8 compares the CSI for heating level, a correlation is obtained based on each specific
the 50 Spanish locations (according to the CEV project) and range of climates, building typologies and construction
the CSI obtained for 15 residential buildings representative practices.
of the Spanish typology and construction, using DOE-2E as
the code for simulation. The quality of the fit demonstrates
USE OF THE CLIMATE SEVERITY INDEX
that the CSI is independent of the way of assessing the
performance of the building. Consequently, it can be applied The aim of the Climate Severity Index (CSI) in the Euro-
to any simulation-based or monitoring-based scheme of class project was to predict the climate-dependent energy
building energy assessment. requirements of a building in a standard year from the
Figure 3.9 shows the effect of trying to apply the results of energy requirements obtained in an actual year.
correlation to a building with a very different typology and Multiplying the heating or cooling requirements
construction standard (the Belgian house referred to as (delivered energy) by the ratio between the reference
building 1 in reference [3]) when placed in the Spanish CSI (superscript N) and the actual CSI produces this
locations. normalization. The following expressions (analogous to
 Energy normalization techniques 67

Equations 3.3 and 3.4) are used: made for periods of:
N
N delivered delivered
CSI heating
Q spaceheating = Q spaceheating · (3.30) • some two weeks (short period)
CSI heating • an extended period much longer than two weeks
N (long period).
N delivered delivered
CSI cooling
Q spacecooling = Q spacecooling · (3.31)
CSI cooling The length of the monitoring period depends on the
purpose of the measurement. The short period is to monitor
the size of the specific energies and to record temperatures.
NORMALIZATION WITH RESPECT TO THE The extended (long) period is primarily for assessing the
INTERNAL ENVIRONMENT CLIMATE heat-loss factor of the building by measuring delivered heat
Normalization with respect to the internal environment and temperatures.
climate requires a set of conditions that specify what the The monitoring period and the choice of normalization
internal climate is and, for the occupants, their rate of method will influence how to determine energy consump-
presence and their habits. In other words, the standard tion with normal internal and external conditions.
occupant has to be defined. The variables and parameters
that will be influenced by standardization, described in a
The heating and cooling degree-day method
generalized manner, are:
The short-term monitoring campaign verifies the
• the temperatures and temperature control of the indoor quantities of specific energy and the internal temperature.
climate This monitored data will not be adequate to directly assess
• energy use for heating standard tap-water consumption the heat-loss factor. However, together with billed quantities
• energy use for a standard set of household appliances and information from the audit, this information can be
and lighting used in the same way as BEP calculations are performed.
• a standardized method of shading control. The difference is that specific energy is more accurately
assessed and temperatures are more precise.
This task, deciding the choice of the set of variables The degree-day concept has been presented above and is
and parameters, including the procedures for determining repeated here. For the heating season
numerical values, should be determined by national
tend
interests. Application of normalization of the internal HDD = (θHDD_base − θe ) (3.32)
tstart
environment is only done within MEP, and the motivation
for this was stated in the section ‘Normalization of BEP and while for the cooling season
MEP output’ earlier in this chapter.
tend
The procedure for normalization of indoor conditions CDD = (θe − θCDD_base ) (3.33)
tstart
should be considered to be an extra service within MEP.
The quantities of energy determined by this procedure will
The internal temperatures θHDD_base and θCDD_base are base
provide to the end-customer (the occupant(s)) information
indoor temperatures for the period considered, and
on how good their thermal behaviour is in comparison to
normalization with respect to the external climate often
that of the standard occupant, had the standard occupant
uses default values for normalizing space-heating and space-
lived in the residence considered. For example, how much
cooling energies. In equation form, these are
more or less energy for tap-water heating is the occupant
using in comparison to a standardized quantity? N delivered delivered HDD N
Within this framework, only normalization of the Q spaceheating = Q spaceheating · (3.34)
HDD
temperature of the indoor environment will be considered.
As stated above, normalizing values for other variables N delivered delivered CDD N
Q spacecooling = Q spacecooling · (3.35)
is left open for national initiatives to complete within CDD
the framework. The primary motivation for normalizing
internal temperatures only is that these directly influence The base temperatures are default values and roughly
space-heating and cooling energy. correspond to the temperatures that would have been
To some extent, normalization of internal environment found if heat gains had been omitted in the building, yet
is performed in the procedures of normalization with the heating system delivered the same quantity.
respect to external climate. This is implicit in the choice of From the MEP point of view, the mean temperature of
internal base temperature in the degree-day method (DDM) the internal environment θi is assessed. Although the moni-
and in relating the heat loss to the set-point temperature toring period is short, the assumption is that this internal
in the modified utilization factor method (MUFM). The climate is constant throughout the season. Therefore, a
Climate Severity Index is based on numerous simulations calculation is made for the actual and the reference years,
where the indoor conditions have been prescribed. such that, for the heating season
The methodology within MEP requires that temperatures tend
in the thermal zones be measured. Measurements are HDD = (θi − θe ) (3.36)
tstart
68 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

while for the cooling season The normalized utilization factor is calculated using the
tend normalized gain–loss ratio, here expressed as
CDD = (θe − θi ) (3.37)
tstart
N ei 1 − (γN ei )a
ηUF = if (γN ei ) = 1 (3.44)
Note especially that θi is completely different from the base 1 − (γN ei )a+1
temperatures used earlier. Whereas the base temperature N ei a
ηUF = if (γN ei ) = 1 (3.45)
refers to the effect of space heating/cooling only, θi also a+1
involves heat gains.
At a national level, the normal mean temperature of the The normalized space-heating energy is calculated from the
N
internal environment θi is prescribed. The degree-days of set of variables as
N ei delivered N ei at_spt N ei delivered N
the reference (normal) heating and cooling periods are Q spaceheating = Q heat losses − ηUF (Q appliances + Q metabolic + Q solarheat )
computed according to (3.46)
N
tend N N
HDDie = (θi − θe ) (3.38) If the set-point temperature cannot be assessed during the
tstart
audit or during the monitoring campaign (for example,
for the normal heating season, and by analysing temperature profiles during the night), the
N
tend N N alternative is to use the lowest recorded temperature during
CDDie = (θe − θi ) (3.39)
tstart the measurement period as the set-point temperature.

for the normal cooling season. The subscript ie denotes

that normalization takes into consideration both reference Normalization with heat-loss factor methods
internal and external climates.
The three procedures of normalization in the previous
The normalization calculations are then made with the
sections involve reference internal and external climates
equations as follows:
and are applied to MEP calculations. Primarily the short-
N term monitoring, together with the use of bills, allows the
N ie delivered delivered
HDDie
Q spaceheating = (Q spaceheating ) · procedures to be applied.
HDD ⎛ ⎞
HDDie
N For the extended monitoring period, which is primarily
+ Q metabolic + Q solarheat ) · ⎝ ⎠
delivered
+ (Q appliances dedicated to determining the heat-loss factor of the build-
HDD − 1
ing, another alternative can be employed. This is by using
simulations. The experimental heat-loss factor, together
(3.40)
N with reference internal and external climates, is used for
N ie delivered delivered
CDDie
Q spacecooling = (Q spacecooling ) · determining the normalized space-heating energy (and
CDD ⎛ ⎞ space-cooling energy).
N
CDDie Below, brief information is given on two methods that
+ Q metabolic + Q solarheat ) · ⎝1 − ⎠
delivered
+ (Q appliances
CDD can be utilized.

(3.41)
THE ENERGY SIGNATURE METHOD

The Energy Signature (ES) method provides an experi-

The modified utilization factor method
mental value for the heat-loss factor, H. With a set of pre-
The modified utilization factor method does not explicitly scribed internal and external temperatures, solar radiation
make use of internal temperatures. Calculations are based and optionally internal heat gains, the space-heating energy
on the set-point temperature of the main heating system. for the normalized conditions can be calculated. A general
When normalization is only with respect to the external formulation of a static model will be such that
climate, the actual set-point temperature θspt is used. N ei N N
 N
Q spaceheating = H(θi − θe ) · t − As,n · q Ng,n − int gains · t
For normalization with respect to internal climate, the
changes will appear in the ‘normal’ (reference) set-point (3.47)
N
temperature, here called θspt . The numerical value for the where
reference set-point temperature is prescribed at a national t = the simulation time step (for example 1 month)
level. N N
θi and θe = environment temperatures for the reference
The six working steps of the methodology are to be year
followed, with the substitution of one variable. The changes As, n = solar apertures in direction n
are listed below. N
q g,n = global radiation onto a surface in direction n for
N N the reference year
N ei at_spt at_spt
(θspt − θe )
Q heat losses = Q heat losses · (3.42) N
int gains = the power of internal gains, possibly with
(θspt − θe )
normal values
delivered N
Q appliances + Q metabolic + Q solarheat
γN ei = (3.43) This method is similar to the MUF method. Whereas the
N ei at_stp
Q heat losses MUF method implicitly calculates a heat-loss factor for each
 Energy normalization techniques 69

data period considered, the ES method makes use of the The following areas are not considered to be a part of the
mean heat-loss factor for all periods. heated floor area:

• rooms lacking heating units and having a space

THE UA AND gA METHOD temperature less than θheated
• heated garages if these are outside the building envelope
The UA and gA method, developed by Somogyi,5 makes use
• passively heated garages if the space temperature is less
of an RC network and a Monte-Carlo approach to identify
than θheated
the UA value of the envelope. The calculation involves a
• external enclosed volumes with passive heating or
simulation of the building subjected to a reference external
occasional heating, such as glazed balconies
climate and a fixed set of internal conditions, see Wouters
• sun-courts without heating units or where the
and Loncour6 and the literature review in Chapter 1.
temperature is less than the value θheated
• the lower horizontal areas of heated or non-heated
NORMALIZATION WITH RESPECT TO crawl spaces
BUILDING SIZE • cellars without active heating systems
• saunas that are used less than twice a week
A way of normalizing energy consumption to obtain a value • storage rooms without heating units and with closed
that is independent of the size of the building is by dividing entrances.
the energy use by the heated floor area or the volume
enclosed by the building. The heated floor area of the The following areas are considered to be a part of the
building is denoted by A. The energy consumption, either heated floor area, although the space to which the floor is
global or specific, is generally defined as exposed to may not be equipped with a heating unit:

Q • spaces that lack heating units but are completely

q= (kWh/m2 · year) (3.48)
A surrounded by heated spaces (completely internal spaces)
• spaces that lack heating units but are equipped with
The definition of the heated or living floor area has a ventilation inlet terminals (for exhaust air)
large impact on the magnitude of the area-specific energy • Halls and corridors with doors that are constantly open to
requirement. adjacent heated spaces
• open staircases (area = projection onto each horizontal
plane).
DEFINITION OF THE FLOOR AREA

Relating energy use to floor area is a common concept.

However, national traditions and applications have different
Area-specific space-heating energy
definitions for the size of floor area. These can vary from Energy for space heating is conveniently related to floor
being the external or internal measure of the perimeter of area:
the building envelope; they may also account for or exclude superscript
Q spaceheating
internal partition constructions, staircases, etc; and they superscript
q spaceheating = (kWh/m2 · year) (3.50)
may take into account whether or not spaces of the exposed A
floor are heated/cooled. where the superscript can be :
Within the framework of this project, and the application
• delivered for area-specific delivered space-heating energy
of the methodology to buildings, the proposed definition of
• N delivered for area-specific normalized delivered
the ‘heated floor area’ is defined as in the following.
space-heating energy
The heated floor area is the sum of areas within the
• supplied for area-specific supplied space-heating energy
building envelope, based on the interior measures along
• N supplied for area-specific normalized supplied
the building envelope, which are exposed to living spaces
space-heating energy.
that have a temperature that exceeds θheated (see definition
below). The floor area includes the areas covered by partition
constructions. Equivalent area-specific space-cooling energy
The definition of θheated is such that
Space cooling is a non-linear phenomenon and, when
utilized, it is seldom that the whole building is cooled for
θheated = θ̄e + 0.7 · (θspt − θ̄e) (3.49)
a long period of time. Moreover, the fact that only some
of the rooms are cooled, if these are cooled at all, gives
where
rise to problems when defining what floor area is actively
θ̄e = the average temperature of the external environment being cooled. Therefore, the same definition is used as for
during the heating season (◦ C) heated floor area, but is called the equivalent area-specific
θspt = is the set-point temperature of the main heating space-cooling energy
system (◦ C). If this value is not available, default superscript
values for indoor temperatures determined at a superscript
Q spacecooling
q spacecooling = (kWh/m2 · year) (3.51)
national level can be utilized. A
70 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

where the superscript can be: • N supplied for area-specific normalized supplied global
energy.
• delivered for area-specific delivered space-cooling energy
• N delivered for area-specific normalized delivered
space-cooling energy
• supplied for area-specific supplied space-cooling energy REFERENCES
• N supplied for area-specific normalized supplied
space-cooling energy. 1. EN ISO 13790: 2002, Thermal Performance of Buildings – Calculation
of Energy Use for Space Heating.
2. NEN 2916, Energiprestatie van utilitels-gebouwen/Energy
Area-specific global energy performance of non-residential buildings. Bepalingsmethode,
Annual area-specific global energy is the sum of specific Nederlands Normalisatie-instituut NEN, Delf, The Netherlands
(in Dutch).
energies, divided by the heated floor area. Expressed in
3. EUROCLASS, 2001, Final report of EUROCLASS: Develop-
equation form, the actual area-specific global energy use is
ment of an European Methodology to Experimentally Assess
formulated such that
and Clasify Existing Residential Buildings Based on
superscript
q global = Their Actual Energy Consumption, Contract No. XVII/
4.1031/Z/99–330, EU-SAVE programe.
superscript superscript superscript superscript superscript
Q spaceheating + Q spacecooling + Q appliances + Q hotwater + Q external 4. Ministerio de Fomento e IDEA, 1999, Fundamentos Técnicos de la
A Calificación Energética de Viviendas.
5. Somogyi Z, 1998, In Situ Evaluation of the Thermal Characteristics
(3.52) of Building Components and Buildings including Comparison with
where the superscript can be: Predicted Performances. PhD thesis UCL, Louvain-La-Neuve.
6. Wouters P and Loncour X, 2001, In Situ Identification of UA and
• delivered for area-specific delivered global energy gA-value: An Overview of Possibilities and Difficulties. Save Euro-
• N delivered for area-specific normalized delivered global class report, Department of Building Physics, Indoor Climate
energy and Building Services, Belgian Building Research Institute,
• supplied for area-specific supplied global energy Brussels.
 71

CHAPTER 4
The Euroclass method – description
of the software
SERVANDO ÁLVAREZ, ANTONIO BLANCO, JUAN ANTONIO SANZ AND FRANCISCO J SÁNCHEZ
Engineering School, DIE-Grupo Termotecnia, University of Serville.
Camino de los descubrimientos s/n. Sevilla, Spain

THE RATING METHODOLOGY EUROCLASS: DEVELOPMENT OF A EUROPEAN

Any rating procedure is a comparison scheme that makes
METHODOLOGY TO EXPERIMENTALLY ASSESS
it possible to give a score to a certain building (the rated
AND CLASSIFY EXISTING RESIDENTIAL
BUILDINGS BASED ON THEIR ACTUAL
building). It is based on three issues:
ENERGY CONSUMPTION CONTRACT
NO. XVII/4.1031/Z/99–330
• the variable of performance or the set of variables that are
going to be compared
• the comparison scenario, that is, the group of buildings
(or the group of values) that are going to provide the
distribution of the variable of performance, creating the
framework of comparison
• the rating score, which includes the criteria and the limits
that give the score when the variable of performance is
compared in the comparison scenario.

A common framework for a rating methodology has been

developed within the framework of the Euroclass project.
It can be widely applied to the different methods (deter- Description of the software
ministic or experimental) of characterization of the energy
This software implements the rating procedure developed
consumption of residential buildings. At the same time, it
for the Billed Energy Protocol (BEP) and for the Monitored
can be adapted to national peculiarities and energy policy
Energy Protocol (MEP), explained in Chapters 2 and 3. The
criteria. The framework is based on the use of the relative
software makes it possible to introduce general data for the
frequency distribution curves for the different end-uses of
building characteristics and the site, as well as from one
the energy. For every similarity level these curves are assumed
of the above-mentioned protocols. Each of these protocols
to be independent of the characterization method.
provides useful information for carrying out a rating test of
The rated variables are: total supplied energy in kWh/m2
the building either in the cooling season or in the heating
and total delivered energy in kWh/m2 for heating and/or
season. The program provides a rating for the building in a
cooling purposes.
specific comparison scenario.
These two variables can mainly be obtained either from
The tool has been developed in Excel workbook format.
the Billed Energy Protocol (BEP) or from the Monitored
It includes four sheets. Two sheets contain the input–output
Energy Protocol (MEP), explained in Chapter 2, although
variables, which ask for the required data and rating choices
a deterministic approach would also be acceptable. For
and present the results. Two further sheets include a database
every case, the rating procedure uses the results specific to
with information to be completed with country-dependent
the approach used.
requirements, and a calculation sheet.
In order to compare buildings within a common context,
two successive normalizations are performed: normalization
with respect to the climatic conditions and normalization
Input and output sheets
with respect to the operating conditions. The input sheet is named ‘Main data 1’. This sheet includes
A simple Excel-based program has been prepared to two windows in which climatic and either heating or cooling
implement the rating methodology. Figure 4.1 shows the data are provided, for BEP or MEP. The output sheet is
output screen of the software mentioned. called ‘Results’ and includes the comparison scenario and
72 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Figure 4.1 The output screen of the software program

the rating results. Actual and normalized energy consump-

tions, as well as the building score, are presented in this
sheet. A more detailed description can be found in the
Figure 4.2 The input screen of the software program
following sections.

The database DISTRIBUTION CALCULATION SET OF VALUES FOR REFERENCES

A third sheet, named ‘Update Database’, includes the The ratio of the 90/10 percentiles, λ (lambda), and the 10%
database, which will probably be edited to provide national inflection percentile.
data for the rating. The data are stored in tables that
correspond to the windows in the input and output sheets.
THE CALCULATIONS SHEET

CLIMATIC DATA SETS The fourth (hidden) sheet is named ‘Calculations’ and must
not be edited. It is the basis for the calculations that are used
Data for about 300 locations can be stored. The name of the
to carry out the rating.
set is required (city (country)) and is shown in the combo
set. 48 values are located after the name, four-monthly sets
for degree-days, n/N, total over horizontal surface radiation Input and output screens
(kWh/m2 ) and outside temperature. The input and output screens are shown in Figures 4.2
and 4.3.
INERTIA RATIOS

Three inertia ratios have been taken into account, light, Description of variables
mean and heavy. These values are calculated as e −1 /τ where DESCRIPTION OF THE INPUT
τ is the main time constant of the building in hours.
Figures 4.4 to 4.9 show areas of the input screen with
annotations describing the different areas.
HEATING REFERENCE CONDITIONS

This set of values comprises the reference conditions against DESCRIPTION OF THE OUTPUT
which internal operation normalization will be carried out.
About 100 values can be filled in. Figures 4.10 to 4.12 show areas of the output screen with
annotations describing the different areas.
COOLING REFERENCE CONDITIONS
DESCRIPTION OF THE DATABASE
These are similar to the heating reference conditions, but
for the cooling-dominant season. Figures 4.13 to 4.15 show areas of the database screen with
annotations describing the different areas.
REFERENCE RATING DISTRIBUTION
Example
About 100 different distributions can be included. Data
required are the 10th percentiles for every distribution set. Figures 4.16 and 4.17 show the input and output screens for
The format for the values is as a percentage. an example application.
 The Euroclass method – description of the software 73

Figure 4.4 An empty input screen of the software program

Figure 4.3 An empty output screen of the software program

Figure 4.5 Area of the input screen. Climatic data

Figure 4.6 Area of the input screen. Heating data 1

74 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Figure 4.7 Area of the input screen. Heating data 2

Figure 4.8 Area of the input screen. Heating and cooling data

Figure 4.9 Area of the input screen. Calculate and Help buttons
 The Euroclass method – description of the software 75

Figure 4.10 Area of the output screen. Comparison scenario

Figure 4.11 Area of the output screen. Rating 1

Figure 4.12 Area of the output screen. Rating 2

76 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Figure 4.13 Area of the database screen. Climatic data

Figure 4.14 Area of the database screen. Reference conditions

Figure 4.15 Area of the database screen. Comparison scenarios

 The Euroclass method – description of the software 77

Figure 4.16 Input screen of the example

Figure 4.17 Output screen of the example
78

CHAPTER 5
Examples and case studies
P. WOUTERS AND X. LONCOUR
Division of Building Physics and Indoor Climate, Belgian Building Research Institute,
Brussels, Belgium

INTRODUCTION Table 5.1 Definition of the categories of building to consider

Type of bills Heating system Heating and cooling system
The methodology in the preceding chapters has been
Measured Single-use Single-use
applied in four countries (Belgium, Greece, Spain and
Multiple-use Multiple-use
Sweden). This chapter describes the results of the
Estimated Single-use Single-use
procedure, the problems encountered and the solutions
Multiple-use Multiple-use
envisaged to solve them.
This chapter is divided into two parts, the first part deals
with the results of the Billed Energy Protocol procedure
(BEP) and the second part with the results of the Monitored
Energy Protocol (MEP).
The two protocols have each been applied in the four
considered countries. The application of the methodology,
the way to implement the monitoring, the choice of the
normalization techniques and the calculations have been
implemented at a national level. The authors who have
applied the different protocols and implemented the calcu-
lations are listed at the end of the chapter.

BUILDING SELECTION – CHARACTERISTICS

OF THE BUILDING ANALYSED
Figure 5.1 The four countries considered
Selection criteria
Different major criteria have been considered in the The type of buildings selected in the
selection of the houses: four countries
The diversity of climate encountered in the four coun-
• The way the energy consumption and the energy bills are tries considered (Figure 5.1) makes it possible to test the
determined: energy bills based on estimation (estimated methodology in very different situations with different
energy consumption) or on a measurement (measured billing systems.
energy consumption). Buildings have been further subdivided according to the
• The possibility of distinguishing within the energy following considerations:
consumption (of gas for instance), the part used for
heating/cooling from the part used for other applications • In Belgium two buildings have been analysed for the
(cooking, hot tap-water production). If it is possible to heating consumption on the basis of bills (estimated
completely distinguish the energy consumption for energy consumption): one building in the ‘single-use’
heating/cooling, we speak about single-use of type of fuel; category, the other in the ‘multiple-use’ category.
if not, we speak about multiple-use of type of fuel. • In Sweden two buildings have been analysed for the
• The final parameter to consider is the presence or heating consumption on the basis of bills (measured
absence of a cooling system. energy consumption): one building in the ‘single-use’
category, the other in the ‘multiple-use’ category.
Following these criteria, the next matrix was set up that • In each of Greece and Spain two buildings have been
defines the type of buildings to select. Four major categories analysed for the heating consumption and cooling
of building were defined; these four categories can be split consumption on the basis of bills (measured energy
into two further sub-categories (single-use/multiple-use); consumption): one building in the ‘single-use’ category,
see Table 5.1. the other in the ‘multiple-use’ category.
 Examples and case studies 79

Table 5.2 Distribution of the type of test between the countries Belgian buildings
considered
THE BELGIAN BILLING SYSTEM
Type of bills Heating system Heating and cooling system
Measured Sweden Spain and Greece In Belgium, for most of the time, measured bills (related
Estimated Belgium to gas, electricity or water consumption) are available once
a year. The bills related to the supply of heating oil are
Table 5.3 Normalization techniques chosen in the four countries available at each delivery. In the majority of cases, only one
considered meter is installed in the building and all the consumption
Country Heating Cooling is measured with this meter. Multi-use of fuel is common.
Belgium Degree-days Not relevant In the multi-family houses, several systems exist with separate
Greece Degree-days/ Climate Severity Index or common measurements.
Climate Severity Index Monthly bills are sent based on the energy consumption
Spain Climate Severity Index Climate Severity Index of the previous year.
Sweden Modified utilization Not relevant
ESTIMATION OF THE UNKNOWN PARAMETERS OF THE
factor method
ENERGY CONSUMPTION

Since multi-use of fuel is the most common system

A total of eight buildings have been analysed. The whole
encountered in Belgium, it will often be necessary to
measurement programme is summarized in the Table 5.2.
estimate different parts of the global energy consumption
(for instance energy consumption for hot water or for
Procedures and normalization cooking). Each method of estimation is based on a number
techniques applied of assumptions.
In the four countries, both procedures (BEP and MEP) have
been applied. The choice of the normalization technique Energy consumption for hot water
to be applied was left to be decided at a national level. As A calculation method applied in Belgium allows the
described in Chapter 3, three techniques are available to assessment of the average hot-water consumption.1 This
normalize the heating/cooling consumption: method is generally used to dimension solar hot-water
systems. The parameters playing a role in the calculated value
• degree-days method (DDM)
are the number of rooms, the localization and the standing
• modified utilization factor method (MUFM)
of the dwelling. This method is used in the scope of these
• Climate Severity Index technique (CSI).
calculations.
For a description of these three normalization techniques, The method gives a figure for hot-water consumption. In
their advantages and limitations please see Chapter 3. order to determine the energy consumption, it is necessary
The techniques have been applied in the different to make assumptions about temperatures and about the
countries as shown in Table 5.3. The normalized results efficiency of the production system.
presented in the next section have been established on the
Energy consumption for cooking
basis of the normalization techniques specified in this table.
In some cases, multi-use of fuel occurs with the energy
consumption for cooking (for instance in the case of gas). It
RESULTS OF THE BILLED ENERGY is therefore necessary to estimate this energy consumption.
PROTOCOL PROCEDURE Information coming from reference [2] makes it possible to
Introduction assess this consumption.
The BEP procedure has already been extensively described
NORMALIZATION – DEGREE-DAYS CORRECTION
in Chapter 2. We recall some aspects here. The Billed
Energy Protocol (BEP) is based on: Several choices can be made for normalizing the
energy consumption with respect to the climate (outdoor
• The collection of energy bills. temperature). Within the scope of the BEP protocol, the
• A set of data collected on site. Data are collected on normalization of the energy consumption has been realized
the type of building, the type of heating/cooling/ by using the degree-days 15/15 concept:
ventilation/hot tap-water systems, the appliances, the
efficiency of the boilers, the set-point temperatures, • The first 15◦ C means that the heating system works when
the behaviour of the users, etc. All these data make the outside temperature is lower than or equal to 15◦ C.
it possible to produce a better analysis of the This first value makes it possible to determine the length
energy bills. of the heating season.
• The climatic data measured in meteorological stations. • The second 15◦ C means that the heating system is shut
down when the indoor temperature reach 15◦ C. This
No special monitoring is realized within the scope of this second value determines the T to take into
protocol. consideration.
80 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

In practice, the duration of the heating season is different centre of Brussels (Figure 5.4). The total heated floor area
in each building; houses that are better insulated with is 273 m2 . Apart from the roof, the house is not insulated.
a higher glazing area will have a shorter heating season There is both single and double glazing.
than other buildings. The choice of a standardized Currently, the building is occupied by two persons
calculation procedure allows the procedure to be applied and is unoccupied for about 20 weeks per year. The
systematically. occupancy pattern has radically changed during recent
Other choices could be made to normalize with respect years. This is clearly shown in the results of the BEP
to the climate (degree-days 15/17). The impact of this procedure.
decision is estimated below. The heating system is a central heating system with gas,
and the hot-water production is provided by the same boiler
THE NORMAL BELGIAN CLIMATE as the heating system. The system is used for hot-water
production in the winter as well as in the summer. The
Figure 5.2 shows the monthly average outdoor temperature
regulation of the heating system occurs via a programmable
for a normal year in Uccle, while Figure 5.3 summarizes the
thermostat situated in the living room. Only one meter for
monthly solar energy received during a normal year on a
the gas consumption is installed in the house. An assessment
horizontal surface.
of the hot-water consumption was necessary. Measured bills
BUILDING NO. 1 are available once per year.
One meter each is also installed for the water
General information and electricity consumption. Measured annual bills are
The first Belgian building selected is a four-storey single- available.
family dwelling built in 1956. The house is located in the No cooling system is installed.

20

18 17.1 16.8
16 15.5
Average outdoor temperature (°C)

14.6
14 12.7
12
10.5
10 9
8
6.1
6 5.5

4 3.5 3.3
2.6
2

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.2 Monthly average outdoor temperature – normal year in Uccle (Brussels)

160,000
Global solar radiation
140,000 Direct solar radiation

120,000
Solar radiation (Wh/m2)

100,000

80,000

60,000

40,000

20,000

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.3 Monthly data for solar radiation – normal year in Uccle (Brussels)
 Examples and case studies 81

Figure 5.4 Belgian building no. 1: street façade (left); garden façade (right)

Available data – calculation and results Table 5.4 Belgian climatic data
Average outdoor temperature
A set of energy bills is available for this house. The procedure Year during the winter (◦ C) Degree-days 15/15
has been applied to the data for the four last years. Table 5.4
Normal year 6.9 2,033
shows the meteorological data for these four years.
1996–1997 6.7 2,067
For this house, the bills are delivered in June between the
1997–1998 8.1 1,658
two heating seasons. The data included in the energy bills
1998–1999 7.5 1,824
are summarized in Table 5.5. In Table 5.6 other information
1999–2000 8.2 1,730
necessary to assess the specific energy consumption is
summarized.
Table 5.7 summarizes the measured and assessed supplied The area-specific results are presented in Table 5.10. The
energy for the specific uses. same error band in the relative values is valid for these
The following set of assumptions is used to assess the hot- results. Figure 5.6 shows these area-specific values.
water consumption: The results for the delivered energy are presented in
Table 5.11.
• hot-water consumption (m3 ) – assessed The fact that the bills are delivered in June makes it easy
• water temperature as supplied (average temperature to use the data to calculate the energy consumption. The
assessed at 10◦ C) and temperature of the hot water degree-days for only one winter are included in the bill.
delivered (set-point temperature equal to 57◦ C) A different normalization procedure could be used, but
• system efficiency (see Table 5.6). the stability of the results would be the same.
Based on these values, it is possible to calculate the supplied A major characteristic of the methodology can be seen
actual and normalized energy consumptions (Table 5.8). here. This is that no normalization is carried out for the
The normalization occurs here according to the 15/15 occupancy pattern, which means that the results presented
degree-days approach. are very dependent on the occupancy. New occupants would
The results are very stable. We can clearly see the change mean that a new analysis of the energy consumption would
of occupancy pattern in the house during the last year be required.
considered. These values are shown on Figure 5.5. An error
band has been calculated by applying a variation of ±3◦ C BUILDING NO. 2
on the temperature.
General information
The numerical values of the error band are presented in
Table 5.9. These values are given in both absolute and relative The second building considered in Belgium is a detached
terms. It can be seen that the accuracy seems to be very good building, which was built in 1985 and is situated in the
(maximum 6% deviation compared to the calculated value). country. The building is shown in Figure 5.7. The total
82 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.5 Supplied energy – based on bills

Number of days Number of Number of
Supplied energy between bills occupants weeks away Gas (kWh) Electricity (kWh)
Estimated (E) or measured (M) M M
Conversion factor 10.28 kWh/m3 1 kWh/kWh
17 June 1996–16 June 1997 364 4 3 39,392 5,964
16 June 1997–12 June 1998 361 4 3 34,074 5,360
12 June 1998–17 June 1999 370 3 5 35,353 5,378
17 June 1999–28 June 2000 377 2 20 24,666 3,970

Table 5.6 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E ), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances + lighting Electricity 100% 100% E
Heating system gas Gas 60% – E
Hot-water system − combined with the heating system Gas 40% 40% E

Table 5.7 Supplied energy (in kWh)

Energy consumption Hot-water Cooking Energy consumption
– gas consumption consumption Qheat – electricity Qtot
Measured (M), estimated (E) M E E C M C
or calculated (C)
Year Number of days
1996–1997 364 39,392 5,372 923 33,097 5,980 45,372
1997–1998 361 34,074 5,416 931 27,726 5,419 39,493
1998–1999 370 35,353 5,069 871 29,413 5,305 40,658
1999–2000 377 24,666 3,387 582 20,697 3,844 28,509

Table 5.8 Supplied actual and normalized energy consumptions Table 5.10 Supplied area-specific actual and normalized energy
Qheat, sup Qheat, sup, normal Qtot, sup Qtot, sup, normal consumption
Year (kWh/year) (kWh/year) (kWh/year) (kWh/year) qAheat, sup qAheat, sup, normal qAtot, sup qAtot, sup, normal
1996–1997 33,097 32,561 45,372 44,836 (kWh/m2 (kWh/m2 (kWh/m2 (kWh/m2
1997–1998 27,726 33,996 39,493 45,763 Year year) year) year) year)
1998–1999 29,413 32,797 40,658 44,042 1996–1997 121 119 166 164
1999–2000 20,697 24,333 28,509 32,146 1997–1998 102 125 145 168
1998–1999 108 120 149 161
1999–2000 76 89 104 118
Table 5.9 Supplied actual and normalized energy consumption – error
band
Year Qheat, sup, normal (kWh/year) Qtot, sup, normal (kWh/year) Table 5.11 Delivered normalized energy (in kWh)
1996–1997 32,561 < Q < 33,067 44,836 < Q < 45,342 Year Qheat, N, del qheat, N, del Qtot, N, del qtot, N, del
(Delta max 1.6%) (Delta max 1.1%) 1996–1997 19,536 71.6 27,665 101.3
1997–1998 33,168 < Q < 35,954 45,763 < Q < 47,720 1997–1998 20,398 74.7 27,984 102.5
(Delta max 5.8%) (Delta max 4.3%) 1998–1999 19,678 72.1 27,011 98.9
1998–1999 32,546 < Q < 33,023 43,791 < Q < 44,268 1999–2000 14,600 53.5 19,798 72.5
(Delta max 0.9%) (Delta max 0.6%)
1999–2000 24,333 < Q < 24,872 32,146 < Q < 32,685
measures the gas consumption for hot water (the readings
(Delta max 2.6%) (Delta max 1.7%)
are taken manually and there has been no systematic record
in recent years). This kind of situation is exceptional in the
heated floor area is 276 m2 . The house is unoccupied for Belgian context. We are have here the use of a single fuel
about two weeks per year. (except for the energy consumption for cooking, which is
The heating system is a central heating system with gas. marginal).
The production of hot water is realized with a separate
boiler and a solar system. Two gas meters are installed in
Available data – calculation and results
the house; the first general meter measures the total gas
consumption of the building and the bills are established The energy bills are available for recent years. The bills are
on the basis of readings from this meter. The second meter delivered once a year in February. It is necessary to split
 Examples and case studies 83

50,000
Qsup, glob, N
45,000
Qsup, heat, N
Normalized consumption (kWh) 40,000

35,000

30,000

25,000

20,000

15,000

10,000

5,000

0
1996-1997 1997-1998 1998-1999 1999-2000
Figure 5.5 Belgian building no. 1: Evolution of the normalized total and heating energy consumption over four years

200
Normalized are-specific energy consumption (kWh/m2 year)

qAsup, glob, N
180 168
164 qAsup, heat, N
161
160

140
125
119 120 118
120

100 89
80

60

40

20

0
1996-1997 1997-1998 1998-1999 1999-2000
Heating season
Figure 5.6 Belgian building no. 1: Evolution of the area-specific normalized supplied total and heating energy
consumption over four years

Figure 5.7 Belgian building no. 2: Pictures of the house

84 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

the energy consumption over the two winters covered by shows the energy consumption normalized with respect to
the bill to establish the energy consumption of the building. the energy consumption of the appliances, together with the
Unfortunately, manual readings of the second gas meter are normalized area-specific energy consumption.
not available. Within the BEP procedure, multiple use of fuel The energy consumption is relatively constant over the
was considered. three years. The specific consumptions are rather low, which
The meteorological data are available and have already can be explained by the good level of insulation of the
been presented in Table 5.4. In Table 5.12, the calculation building and the building design developed to increase the
carried out to split the consumption between the two winters solar gains.
is summarized, while the data taken from the bills are The results are presented with an error band in Table 5.17.
summarized in the Table 5.13 and the efficiency of the These results have been calculated by applying a variation
different systems in Table 5.14. of ±3◦ C in the DD correction. The values are given in both
The main information coming from the bills is sum- absolute and in relative terms. The results are represented
marized in Table 5.15. An assessment of the contribution graphically in Figures 5.8 and 5.9.
of the solar system to the energy consumption for hot Based on the efficiency values given in Table 5.14,
water has been made (50%). On this basis, the energy Table 5.18 gives the values of the delivered energy, while the
consumption for heating has been calculated. Table 5.16 area-specific results are presented in Table 5.19.

Table 5.12 Calculations carried out to split the energy consumption between the two winters
Number of days between Number of days in Number of days in Degree-days in Degree-days in
Year the two bills winter 1 winter 2 winter 1 winter 2
1997–1998 360 101 125 564 1,218
1998–1999 374 86 144 579 1,439
1999–2000 357 75 136 503 963

Table 5.13 Supplied energy, based on bills

Number of days Number of Number of
Supplied energy between bills occupants weeks away Gas (kWh) Electricity (kWh)
Estimated (E) or measured (M) M M
Conversion factor 10.32 kWh/m3 1 kWh/kWh
17 February 1997–12 February 1998 360 4 2 12,484 3,750
12 February 1998–21 February 1999 374 4 2 16,156 3,780
21 February 1999–20 February 2000 357 4 2 12,504 3,849

Table 5.14 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E ), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances + lighting Electricity 100% 100% E
Heating system gas Gas 75% – M
Hot-water heating system −
gas (solar pre-heated hot water) Gas 75% 75% E

Table 5.15 Supplied energy (in kWh)

Energy consumption Hot-water Contribution of Cooking Energy consumption –
– gas consumption solar system consumption Qheat electricity Qtot
Measured (M), estimated (E) M E E E C M C
or calculated (C)
Year Number of days % solar system
1997–1998 360 12,484 2,133 50% 1,360 8,991 3,750 16,234
1998–1999 374 16,156 2,216 50% 1,413 12,527 3,780 19,936
1999–2000 357 12,504 2,115 50% 1,349 9,040 3,849 16,353
 Examples and case studies 85

Table 5.16 Normalized and area-specific supplied energy electrical entities is relative simple, there may be problems
QN, heat, sup qN, heat, sup QN, tot, sup qN, tot, sup with non-billed fuels. In rural areas, it is common for biomass
Year (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year) fuels to be gathered locally and combusted for the purpose of
1997–1998 10,258 37.2 17,602 63.7 heating, in stoves, fireplaces and boilers. The only records are
1998–1999 12,621 45.7 19,851 71.9 usually what the owner/end-user remembers. Furthermore,
1999–2000 12,542 45.4 20,019 72.5 since the mid 1990s, the government has issued subsidies for
installing biomass stoves.
Multi-family buildings are commonly heated by means of
Table 5.17 Supplied actual and normalized energy consumption – error district heating. A major problem here is that most buildings
band are not equipped with devices that monitor the consumption
Qheat, sup, normal Qtot, sup, normal within individual apartments. It is very common for all
Year (kWh/year) (kWh/year) energy costs, apart from those for electricity for domestic ap-
1997–1998 10,258 < Q < 10,701 17,602 < Q < 18,044 pliances and lighting, to be shared by residential tenants/
(Delta max 4%) (Delta max 2.5%) occupants on the basis of floor area. With an infrastructure
1998–1999 12,486 < Q < 13,193 17,716 < Q < 20,424 that lacks meters for each apartment, MEP is likely to be the
(Delta max 4.5%) (Delta max 2.9%) protocol that is applied for these buildings.
1999–2000 12,287 < Q < 14,268 19,764 < Q < 21,744
(Delta max 13.8%) (Delta max 8.6%)
CLIMATE DATA

Swedish buildings The Swedish Meteorological and Hydrological Institute

(SMHI) collects climate data. With 372 sites across Sweden,
THE SWEDISH BILLING SYSTEM
the availability of climate data is high. However, not all
The Swedish billing system is usually based on monthly bills. stations provide data on solar radiation and the sampling
The billed quantities are either measured values or predicted frequency may differ depending on the site. For the purpose
values based on earlier measurements. Readings are made of the experimental on-line protocol, apart from solar
(or should be made) at least once per year. However, there is radiation, temperature measurements and degree-day data
a change taking place, where more sophisticated monitoring are available on a month-wise basis.
equipment is being installed to make fuel debits based on For Sweden, there is a so-called ‘normal year’ for various
real monthly consumption. The fast development of on-line regions. The normal year is the average of 30 years,
systems or a tele-infrastructure makes it possible for these represented by data for 1961–1990. Global solar data is
types of services to be introduced. available for the normal year, but not for the diffuse and
As a result of national energy policies after the oil direct radiation onto a normal (perpendicular) surface. For
crisis in the 1970s, many single-family dwellings (47%) are this reason, data has been used for the period 1983–1998.3
equipped with electrical heating. The wider deployment of It was not until 1983 that diffuse and direct radiation were
the experimental on-line collection protocol should make continuously measured at several sites.
it possible to monitor the multipurpose use of electricity. Normalization of external climate is based on the normal
Monitoring equipment that makes use of a building’s central year. The degree-day data provided by SMHI assumes a
distribution unit is convenient. Although measurement of base temperature of 17◦ C. By default, the heating season

25,000
Normalized supplied endrgy (kWh/year)

20,000

15,000

10,000

5,000

0
BEP −1997−1998 BEP −1998−1999 BEP −1999−2000
Figure 5.8 Belgian building no. 2: Evolution of the normalized total and heating energy consumption over three years
86 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Normalized supplied specific energy consumption (kWh/m2 year)

90

80
72 73
70
64

60

50 46 45

40 37

30

20

10

0
BEP–1997–1998 BEP–1998–1999 BEP–1999–2000
Figure 5.9 Belgian building no. 2: Evolution of the normalized area-specific total and heating energy consumption over three
years

Table 5.18 Delivered normalized energy (in kWh)

Hot-water Cooking Energy
consumption consumption Qheat, N consumption – electricity Qtot, N
Measured (M), estimated (E) E E C M C
or calculated (C)
Year
1997–1998 3,374 1,379 7,694 3,802 16,248
1998–1999 3,374 1,379 9,465 3,689 17,907
1999–2000 3,374 1,379 9,407 3,935 18,094

Table 5.19 Delivered normalized energy (in kWh)

changed the fuel usage. From then on, the electrical water-
Year Qheat_N_del Qheat_N_del Qtot, N, del qtot_N_del
based heating system was used as the base heating system,
1997–1998 7,694 27.9 16,248 58.8
with a constant set-point temperature of 17◦ C. The heating
1998–1999 9,465 34.3 17,907 64.9
system is operated all year round, while the fan stove is
1999–2000 9,407 34.1 18,094 65.5
used as a complementary unit. Wood is gathered from
the grounds of the summer cottage, so there are no bills
is considered to last from 15 September until 15 May; the for this fuel. Nor are there any detailed records of wood
summer period starts when the average daily temperature is consumption. The occupier has therefore estimated the
higher than 10◦ C. use of wood on a seasonal (annual) basis, partly based on
Figure 5.10 shows the monthly average outdoor tem- knowledge of the global energy use prior to the installation
perature in Sweden for the normal year, while Table 5.20 of the fan stove and partly on the basis of the volume of
gives some representative data. the shed where the wood is stored. Prior to the installation
of the stove, electricity bills indicated a usage of about
BUILDING NO. 1 20,000 kWh/year but this was for four persons in the
apartment.
General information
Supplied electricity is monitored on an hourly basis. On
The apartment analysed is situated in a semi-detached two- a daily basis, the supplier collects data at midnight over a
family house built in 1982, subject to Swedish building telephone connection. Monitored values are actually listed
code SBN 80. The building contains two apartments in all. on the bill, which is sent to the customer at the start of
The building is relatively well insulated and the mechanical the following month. Within the energy audit of BEP, the
ventilation is equipped with a heat exchanger. Situated monthly summed values have been used.
next to a hill in a suburban area, the building is generally Formally, it is not evident that BEP can be applied in
shaded from the sun. Figure 5.11 shows pictures of the this case because of the uncertainty about the amount of
building. wood concerned. However, within the framework of the
Until 1993, the only fuel type was electricity development of the methodology, it was interesting to
(multipurpose). The installation of a wood fan stove in 1993 evaluate this case using BEP. MEP application will provide
 Examples and case studies 87

20
18 16.7
15.7
16 15.1

14
Average outdoor temperature (°C)

12 11.3
10.4
10
8 7.0
6
4.2
4
1.8
2
0
-2 -0.5
-1.9
-4
-3.7 -3.8
-6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.10 Monthly average outdoor temperature in Stockholm, Sweden for a normal year

Table 5.20 Climate data variables for mean temperature, direct solar radiation onto a normal surface, diffuse solar radiation and the global solar radiation
onto a horizontal surface for a normal year
Actual year Normal year Actual year Normal year
Mean Mean Direct solar Diffuse solar Global solar Direct solar Diffuse solar Global solar
temperature temperature radiation radiation radiation radiation radiation radiation
Month (◦ C) (◦ C) (Wh/m2) (Wh/m2) (Wh/m2) (Wh/m2) (Wh/m2) (Wh/m2)
December 1999 −1.6 −1.9 16,484 4,860 5,900 13,600 4,500 6,000
January 2000 −1.2 −3.7 33,285 7,661 12,080 18,300 7,400 10,000
February 2000 −0.1 −3.8 48,482 16,842 28,040 38,000 16,400 25,600
March 2000 1.6 −0.5 138,989 31,982 80,640 68,400 34,400 59,800
April 2000 6.7 4.2 96,270 51,972 95,770 103,700 54,500 104,400
May 2000 12.2 10.4 150,266 75,099 153,190 169,000 68,500 160,900
June 2000 14.3 15.1 137,121 87,221 161,920 146,800 76,000 158,800
July 2000 16.3 16.7 102,765 69,320 124,400 156,700 72,500 160,000
August 2000 15.9 15.7 131,930∗ 34,670∗ 99,050∗ 121,800 61,400 123,100
September 2000 11.1 11.3 150,235 36,016 95,540 92,700 37,400 75,200
October 2000 10.5 7.0 18,314 20,890 26,030 49,800 21,300 35,600
November 2000 6.8 1.8 3,286 5,965 6,520 21,300 8,400 12,000
December 2000 2.6 −1.9 1,551 4,315 4,480 13,600 4,500 6,000
∗ Not available; reference year 1971 data used for August.

more detail on the daily consumption of wood, and should Available data – calculation and results
therefore be more reliable. How large will the differences in The information that is available from the bills and entered
results be? Another problem that this type of consumption in the form tables is summarized in Tables 5.21 to 5.24, with
poses is: should it be considered to be climate-dependent or the system efficiencies of the different systems and how these
climate-independent? Wood is only used during the winter, were assessed given in Table 5.22.
but the increase in consumption as the weather gets colder Table 5.24 lists the periods for which estimations on
is not known. space heating have been made. These correspond to each
Water consumption is not explicitly available. The official month starting at December 1999 and ending at November
meter registers the total consumption of 18 households 2000. The periods correspond to those of the electrical
and the costs are shared among them on an equal basis. bills, but the wood consumption has been distributed
The occupier installed a private water meter in September among the months of the heating season in proportion
1993, but has not made any recordings since then. The first to the temperature difference between the internal and
registered reading was made during the audit. the external climates. Derived values for delivered specific
Monthly climate data was obtained for the year 2000 from energy are entered in the next column. The MUF method
a climate station in the region of Bromma (some 3 km away) has been used for normalizing the space heating. Data is
and solar radiation from a station situated some 10 km away. entered on a month-wise basis in the preceding columns,
88 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Figure 5.11 Swedish building no. 1: Pictures of the house. The street façade is on the left and the garden façade on the right

Table 5.21 Form table 1: Supplied energy from the processing of data are listed in the column
Supplied energy Electricity Wood headed normal space heating in Table 5.24. In the last
Estimated (E) or measured (M) M E by occupant column, a parameter analysis gives the minimum value
Conversion factor 1.00 kWh/kWh 1,500 kWh/m3 for the delivered space heating to be 10,023 kWh. This
December 1999 1,874 was obtained for a time constant corresponding to 125%
January 2000 1,847 of the nominal time constant, a set-point temperature
February 2000 1,459 that is 19−3◦ C and solar apertures 50% larger than the
March 2000 1,426 assessed values. The maximum value was obtained for a
April 2000 1,040 time constant corresponding to 75% of the nominal time
May 2000 688 constant, a set-point temperature that is 19 + 3◦ C and solar
June 2000 509 apertures 50% less than the assessed values. Note that the
July 2000 474 set-point temperature in this calculation is 16◦ C, which is
August 2000 438 lower than the set-point temperature of the main heating
September 2000 640 system.
October 2000 769 The final results of the procedures are shown in
November 2000 963 Tables 5.25 and 5.26.
Heating season December – 3,000
1999–November 2000 Remarks
Sum (kWh/year) 12,127 3,000
Although supplied electricity is measured in detail (hourly
values available), the use of non-billed wood gives rise to
problems when allocations of specific energy are to be cal-
first for the actual year and then for the normal year. The culated. The approximate estimation of wood consumption
data needed are the dry-bulb temperature for the external gives rise to quite unreliable results. This is one of the reasons
environment, direct solar radiation onto the normal plane that either the quantity of non-billed fuels should be verified
and diffuse solar radiation onto the horizontal plane. by a third party or MEP should be applied. The distribution
For the building under examination, the nominal time of wood use over the heating season was in this case assumed
constant was estimated to be 30 hours. Although the set- to be climate dependent. This assumption may be erroneous.
point temperature of the main heating system (the water- The normalization technique used was the MUF method.
based radiator system) is 17◦ C, wood is burnt almost every Although the set-point temperature of the main heating
day during the heating season. For this reason, the average system was known, the use of wood burning with the
temperature (18–20◦ C) given by the occupant has been purpose of heating the building makes the latter set-point
used, corresponding to approximately 19◦ C. The values temperature void. The real set-point temperature, which

Table 5.22 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E), measured (M) or
Energy conversion system Fuel type Winter efficiency Summer efficiency calculated (C)
Household appliances + lighting Electricity 1.00 1.00 E
Electrical boiler for space heating and hot water, Electricity 1.00 1.00 E
10.5 kW (1982)
Fan stove, wood manual feeding (installed 1993) Wood 0.85 0.85 E
 Examples and case studies 89

Table 5.23 Form table 3: Allocation of delivered energy, on an energy conversion system level, for each of the specific energies
Method Footnote 3 Footnote 2 Footnote 1
Electrical conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
December 1999 1,317 380 177 0
January 2000 1,290 380 177 0
February 2000 956 344 159 0
March 2000 869 380 177 0
April 2000 501 368 171 0
May 2000 131 380 177 0
June 2000 0 338 171 0
July 2000 0 297 177 0
August 2000 0 262 177 0
September 2000 101 368 171 0
October 2000 212 380 177 0
November 2000 424 368 171 0
Sum 5,801 4,248 2,079 0
Method Footnote 4
Wood fan stove Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
December 1999 379
January 2000 479
February 2000 450
March 2000 405
April 2000 271
May 2000 126
June 2000 0
July 2000 0
August 2000 0
September 2000 0
October 2000 171
November 2000 268
Sum 2,550
Sum (all) kWh/year 8,351 0 4,248 2,079 0
1. Electricity delivered for tap-water heating is estimated. The energy use is estimated using a ‘standard’ value in Sweden, corresponding to
1,000 kWh/person per year.
2. A base load is estimated on the basis of energy use during the summer period. The mean value of electricity supplied during the period May–September
is used, giving a load of 748.6 W for the winter season (some 18 kWh/day). From this value, the mean power for tap-water heating is subtracted. For the
summer months, appliance energy is obtained by subtracting hot-water energy from delivered electricity.
3. The space-heating energy is estimated as the total delivered electricity minus the electricity for tap water (footnote 1) and for electrical appliances
(footnote 2).
4. The total amount of wood was estimated by the occupant to be 2 m3. The corresponding energy was distributed over the winter months by weighting
according to the temperature difference between indoors and outdoors.

must be considered to be subjective (when the occupant feels BUILDING NO. 2

the need for an increased internal temperature) cannot be
General information
read and will vary over the day.
Solar irradiation was on average 7% of the total delivered Pictures of the building being analysed are given in
heat during the heating season. A minimum of around 1% Figure 5.12. The apartment being considered is situated in
was found in December. Since the solar radiation is very the middle of the roof of the building.
small, the concept of degree-day normalization (DD) can The apartment is situated in one of three buildings within
be applied. In Table 5.27, the new normalization parameters the estate. The estate is owned by a co-operative-living
have been entered. The base temperature in Sweden is, by society, and each occupant indirectly owns the apartment
default, 17◦ C. they live in. Built in 1998 under Swedish building code
The results from application of the DD method give a BBR94, the buildings are equipped with individual metering
value that is 3% less than the value obtained from the MUF of each apartment; this is not the usual method of billing
method. A difference of 312 kWh is noted. The value given within Swedish co-operative-living societies. Traditionally,
by DD lies within the fuzzy band of MUF. occupants pay for the electricity used within the apartment,
90 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.24 Form table 4: Display of chosen processing periods, actual delivered space-heating energy for the periods, numerical values of influencing
variables, and normalized space-heating energy. The bottom two rows indicate the band (minimum and maximum) values that are obtained with
prescribed conditions according to the normalization methods. The method used for normalization was MUF
Space Actual exterior Actual direct Actual diffuse Normal exterior Normal direct Normal diffuse Normal space
Period heating temperature solar radiation solar radiation temperature solar radiation solar radiation heating
December 1999 1,766 −1.6 16,484 4,860 −1.9 13,600 4,500 1,802
January 2000 1,730 −1.2 33,285 7,661 −3.7 18,300 7,400 2,020
February 2000 1,372 −0.1 48,482 16,842 −3.8 38,000 16,400 1,740
March 2000 1,248 1.6 138,989 31,982 −0.5 68,400 34,400 1,523
April 2000 769 6.7 96,270 51,972 4.2 103,700 54,500 1,021
May 2000 279 12.2 150,266 75,099 10.4 169,000 68,500 454
June 2000 0 14.3 137,121 87,221 15.1 146,800 76,000 5
July 2000 0 16.3 102,765 69,320 16.7 156,700 72,500 4
August 2000 0 15.9 131,930∗ 34,670∗ 15.7 121,800 61,400 8
September 2000 101 11.1 150,235 36,016 11.3 92,700 37,400 107
October 2000 397 10.5 18,314 20,890 7.0 49,800 21,300 698
November 2000 690 6.8 3,286 5,965 1.8 21,300 8,400 1,121
Sum 8,351 10,503
Minimum 10,023
Maximum 12,274
∗ Not available; reference year 1971 data used for August.

Table 5.25 Delivered energy; normalized values were obtained with MUF
Area-specific space Area-specific space
Space heating heating Space cooling cooling Global Area-specific global
Year (kWh/year) (kWh/m2 per year) (kWh/year) (kWh/m2 per year) (kWh/year) (kWh/m2 per year)
December 1999–November 2000 8,351 70.2 0 0 14,677 123.3
Normal 10,503 88.3 0 0 16,829 141.4

Table 5.26 Supplied energy; normalized values were obtained with MUF
Area-specific Area-specific space Area-specific
Space heating space heating Space cooling cooling Global global
Year (kWh/year) (kWh/m2 per year) (kWh/year) (kWh/m2 per year) (kWh/year) (kWh/m2 per year)
December 1999–November 2000 8,801 73.9 0 0 15,127 127.1
Normal 11,069 93.0 0 0 17,395 146.2

Table 5.27 Form table 4: Display of chosen processing periods, actual delivered space-heating energy for these periods, numerical values of influencing
variables, and normalized space-heating energy; the method used for normalization is DD
Period Space heating Actual degree-days Normal degree-days Normal space heating
December 1999 1,766 578 579 1,769
January 2000 1,730 565 632 1,935
February 2000 1,372 496 593 1,640
March 2000 1,248 478 547 1,428
April 2000 769 300 384 984
May 2000 279 59 148 700
June 2000 0 8 7 0
July 2000 0 0 1 0
August 2000 0 0 2 0
September 2000 101 153 123 81
October 2000 397 184 305 658
November 2000 690 307 443 995
Sum 8,351 10,191
Minimum –
Maximum –
 Examples and case studies 91

Figure 5.12 Swedish building no. 2: Pictures of the building. The street façade is on the left and a side view on the right

and the rest of the energy bills are shared and included in been recorded) and these values are saved. The measured
the rent. variables are:
With this kind of co-operative-living society, there are
several players involved in the assessment of data: • space heating that is dissipated by the apartment
radiators
• consumption of cold tap water
• The board of the society must agree that a third party
• consumption of hot tap water
can access the data. However, data for an individual
• electricity for household appliances
household can only be given to a third party if the
• space heating and electricity used in common spaces and
individual occupants agree to this. The individual
for common appliances; this also includes bathroom
occupant can refuse delivery of data even though the
towel heaters in the apartments (not individually
board has consented.
measured).
• The occupant cannot themselves provide the data
without the permission of the board. However, the
board can only be given permission to release data if a Water-based system and radiators deliver heat into the
majority of the members have formally voted and apartment. The source is district heating, both for space
consented to this. heating and for tap-water heating. Electricity is used for
• Only the board, and not the individual member, can household appliances and lighting only. Energy delivered
grant the maintenance operator permission to release for the purpose of tap-water heating is included in the
the data. debiting system and is calculated by multiplying the quantity
• In this case, there are two maintenance operators. consumed by the specific heat capacity of the water and a
One is the ordinary maintenance operator (who in this temperature increase of 55◦ C.
case was the building entrepreneur) and the other Use of energy for common spaces and appliances is
specifically deals with the monitoring and control debited on basis of the apartment size (floor area) as
system. a proportion of the sum of the total floor area of all
the apartments. This energy is used for common space
heating (halls, corridors and storage rooms) and for lighting,
With this situation, it is evident that the entire classification elevators, fans and laundries (whether or not the occupants
process must have all the legal aspects cleared at an early use these common laundries).
stage. The apartment considered is situated on top of one of
The monitoring system was installed and run by Landis- the buildings, rather like a villa placed on the roof. There
Staefa for Siemens. Sampling frequencies can be made are two occupants, one of whom is regularly at home. The
every 2 seconds, but the aim was to make monthly debits heated floor area is 106 m2 . Although the apartment is
for energy use (for this reason the temperatures have not equipped with a fireplace, this is never used.
92 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.28 Form table 1: Supplied energy

Supplied energy Electricity (appliances) District heating 1 District heating 2 District + electricity
Estimated (E) or measured (M) M M M M
Conversion factor 1.00 kWh/kWh 1.00 kWh/kWh 1.00 kWh/kWh 1.00 kWh/kWh
January 2000 573 2,497 197 656
February 2000 503 2,135 169 589
March 2000 482 1,830 186 582
April 2000 410 949 137 463
May 2000 406 229 202 152
June 2000 369 66 175 230
July 2000 302 0 164 111
August 2000 300 14 164 171
September 2000 329 158 148 208
October 2000 423 575 180 319
November 2000 399 1,296 142 356
December 2000 490 1,760 208 447
Sum (kWh/year) 4,986 11,509 2,072 4,284

Table 5.29 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Winter Summer Estimated (E), measured (M)
Energy conversion system Fuel type efficiency efficiency or calculated (C)
Household appliances + lighting Electricity (appliances) 1.00 1.00 E
District heating 1 + 2; 1 = space heating, 2 = hot water District heating 1.00 1.00 E
District heating + electricity of external spaces and appliances District heating + electricity 1.00 1.00 E

Available data – calculation and results beneficial for BEP. Delivered energy is directly measured
and recorded. The protocol is easy to apply within the
The information given in Tables 5.28–5.31 is available from framework of apartments with individual metering, and
the bills. this is practised, for example, in all of Germany and
For the current building, the nominal time constant was Switzerland.
estimated to be some 30 hours. The set-point temperatures Although the monitoring system in this case is equipped
are 21.5◦ C during daytime and 19◦ C at night. In the with the option of recording the temperature where the
calculations, the mean temperature is used. The minimum central thermostats are installed, this option has not been
value for delivered space heating is 14,200 kWh, obtained for used. If it had been, the quality of the data is such that the
the time constant corresponding to 125% of the nominal heat-loss factor could have been estimated within BEP. Since
time constant, a set-point temperature that is the nominal the set-point temperatures are known from the audit, these
less 3◦ C and solar apertures that are 50% larger than the could be used as base temperatures.
assessed values. The maximum value of 16,067 kWh was As an optional service, estimation of the heat-loss factor of
obtained for the time constant corresponding to 75% of the the building is possible if the quality of data from supplied
nominal time constant, a set-point temperature that is the energy bills is judged to be adequate. This was done for the
nominal plus 3◦ C and solar apertures that are 50% less than considered apartment. Two methods have been tested:
the assessed values.
Actual and normalized values are displayed in Tables 5.32 • The sum of the delivered space-heating energy and the
and 5.33. gains is assumed to equal the heat loss at a constant
internal temperature. The temperature is assumed to be
the mean value of the set-point temperatures, 21.5◦ C
Remarks (daytime) and 19.0◦ C (night-time).
Three numerical values from the assessment are interesting. • The sum of the delivered space-heating energy and part
The first is the energy for heating tap water. In Sweden, the of the gains, determined on the basis of the utilization
default value of 1,000 kWh/person per year is often used. factor according to the modified utilization factor (MUF)
This corresponds well with the current value of 2,072 kWh method, is assumed to equal the heat loss at the mean
for two persons. Second, the use of appliances and lighting set-point temperature, as above. The time constant of the
was 4,986 kWh (a normal value for Swedish circumstances). building was assumed to be 30 hours.
This is of the same order as energy use for external space
heating and appliances, which corresponds to 4,284 kWh. The mean rate of heat loss is calculated on a monthly basis,
For this case, external energy use may not be neglected. and regressed against the difference between the set-point
Although the legal aspects were complex, as discussed temperature and the external temperature. All monthly data
above, the technical aspects of the billing system as such were are included in the regression. The results are plotted in
 Examples and case studies 93

Table 5.30 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1
Household appliances + lighting Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January 2000 573
February 2000 503
March 2000 482
April 2000 410
May 2000 406
June 2000 369
July 2000 302
August 2000 300
September 2000 329
October 2000 423
November 2000 399
December 2000 490
Sum 4,986
Method Footnote 2 Footnote 3
District heating 1 + 2 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January 2000 2,497 197
February 2000 2,135 169
March 2000 1,830 186
April 2000 949 137
May 2000 229 202
June 2000 66 175
July 2000 0 164
August 2000 14 164
September 2000 158 148
October 2000 575 180
November 2000 1,296 142
December 2000 1,760 208
Sum 11,509 2,072
Method
District heating + Electricity Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January 2000 656
February 2000 589
March 2000 582
April 2000 463
May 2000 152
June 2000 230
July 2000 111
August 2000 171
September 2000 208
October 2000 319
November 2000 356
December 2000 447
Sum 4,284
Sum (all) kWh/year 11,509 0 4,986 2,072 4,284
1. Electricity delivered to the apartment is measured and billed at the end of each month.
2. Delivered space heating is measured for the apartment and billed at the end of each month.
3. The quantity of tap water, hot and cold separately, is measured. The system operator calculates a temperature increase for hot water corresponding
to 55◦ C times the specific heat capacity of water. This quantity is billed.
4. External energy use (measured) for common-space heating, lighting, laundries and operation of installations (fans, pumps and elevators).
94 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.31 Form table 4: Display of chosen processing periods, actual space-heating energy for the periods, numerical values of the influencing variables,
and normalized space-heating energy. The bottom two rows indicate the band (minimum and maximum) values that are obtained with the conditions
prescribed according to the normalization method of MUF
Space Actual Actual Actual Normal external Normal Normal Normal
Period heating external temp direct solar diffuse solar temperature direct solar diffuse solar space heating
January 2000 2,497 −1.2 33,285 7,661 −3.7 18,300 7,400 2,940
February 2000 2,135 −0.1 48,482 16,842 −3.8 38,000 16,400 2,722
March 2000 1,830 1.6 138,989 31,982 −0.5 68,400 34,400 2,299
April 2000 949 6.7 96,270 51,972 4.2 103,700 54,500 1,206
May 2000 229 12.2 150,266 75,099 10.4 169,000 68,500 325
June 2000 66 14.3 137,121 87,221 15.1 146,800 76,000 361
July 2000 0 16.3 102,765 69,320 16.7 156,700 72,500 41
August 2000 14 15.9 131,930∗ 34,670∗ 15.7 121,800 61,400 140
September 2000 158 11.1 150,235 36,016 11.3 92,700 37,400 183
October 2000 575 10.5 18,314 20,890 7.0 49,800 21,300 926
November 2000 1,296 6.8 3,286 5,965 1.8 21,300 8,400 1,931
December 2000 1,760 −1.6 16,484 4,860 −1.9 13,600 4,500 2,346
Sum 11,509 14,932
Minimum 14,200
Maximum 16,067
∗ Not available; reference year 1971 data used for August.

Table 5.32 Delivered energy. Normalized values were obtained with MUF
Space heating Area-specific space heating Space cooling Area-specific cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 11,509 108.6 0 0 18,567 175.2
Normal 14,932 140.9 0 0 21,990 207.5

Table 5.33 Supplied energy. Normalized values were obtained with MUF
Space heating Area-specific space heating Space cooling Area-specific cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 11,509 108.6 0 0 22,851 215.6
Normal 14,932 140.9 0 0 26,274 247.9

Figure 5.13. The conclusions are as follows: In the case when only space heating is regressed against
◦
the temperature difference (excluding May to August, when
• For large temperature differences (>10 C), the
space-heating energy is negligible), the heat-loss factor is
utilization factor is high, which in the calculations gives a
223 W/K. This value is close to the one obtained for the
good agreement between the methods. There is also
50-hour time constant.
good agreement for the heat-loss factor in these regions.
Normalization by the degree-day method (DD) was tested
• For small temperature differences (<10◦ C), the MUF
to study deviations compared to the MUF method. Relevant
method gives a good extrapolation of the wide range of
values are shown in Table 5.35. The difference between the
temperature difference.
results of the two methods is 420 kWh. The DD method
• The results from the two methods diverge for small
gives results 3% lower incomparison with the value obtained
temperature differences (<10◦ C) because of the
with MUF. Almost half of this deviation is because the MUF
influence of the utilization factor. This region, <10◦ C,
method gives a heating requirement during the summer
is defined in Sweden as the summer season, and the
months; see for example August. Furthermore, the solar
traditional ES method would exclude this data in
radiation for August 2000 was not available.
heat-loss factor calculations. However, when the ‘raw
data’ was inspected, only July was without delivery of
space heating. Nevertheless, the delivered heat is small in Spanish buildings
the period May to September, accounting for the five INTRODUCTION
leftmost points for each method.
In Spain, the energy billing system includes natural gas,
The heat-loss factor was 186 W/K for the Energy Signature electricity and fuel-oil use. Electricity bills and consumption
(ES) method using data for all months. From the MUF measurements are distributed every two months by the
method, the corresponding figures were 232 W/K. Table 5.34 Spanish utilities to all residences. Usually, there is a single
shows the result of variations in the time constant. meter per household for all electricity uses. On some
 Examples and case studies 95

4,500 occasions there may be a second night-rate electricity

4,000 meter.
Natural gas comes from the gas piping network. It is
3,500
billed every two months with an indication of the actual
consumption.
Rate of heal loss [W]

3,000
Diesel oil is billed with the purchase of fuel (once or twice
2,500
in a year). For a multi-storey apartment building, the fuel
2,000 bills are distributed every month by the building manager,
together with the other common-use charges. In new apart-
1,500
ment blocks the occupants of each apartment have the capa-
1,000 bility to turn on and off the heating according to their needs
and each flat has its own metering device for billing purposes.
500
Central heating systems are mostly of the hot water/radiator
0 type. There is a central boiler room and a fuel tank.
0 2 4 6 8 10 12 14 16 18 20 22
Temperature difference [K] Although there are no official reference meteorological
Figure 5.13 The heat-loss factor is determined from a linear regression years, there are degree-days and design conditions for sizing
of power versus temperature difference. The stars show the estimated heating and cooling equipment. In forthcoming regulations,
heat loss at the set-point temperature as calculated by means of the synthetic meteorological years are used for 50 locations.
MUF method; linear regression gave the solid line. The circles show the These years have been obtained from monthly average
estimated heat loss that is equal to estimated delivered energy; linear values of the climatic variables obtained for the period
regression gives the dashed line 1961–1990 as published by the World Meteorological
Organization (WMO).
The two buildings analysed and monitored are situated in
Seville. Tables 5.36 to 5.38 and Figures 5.14 and 5.15 give the
Table 5.34 Dependency of the heat-loss factor on the time constant. The relevant climatic variables.
Energy Signature (ES) method is, independent of the time constant. In
terms of MUF, the ES method can be considered as a special case of MUF
where the utilization factor is constant with a value of 1 Table 5.36 Heating degree-days at 20◦ C for the heating season of Seville
Method MUF ES Reference Year
Time constant (h) Heat-loss factor (W/K) Heat-loss factor (W/K) January 276.2
30 232 186 February 210.6
40 227 186 March 170.8
50 222 186 November 155.3
December 261.7
Season 1,074.6

Table 5.35 Form table 4: Display of chosen processing periods, actual

Table 5.37 Heating degree-days at 20◦ C for the cooling season of Seville
delivered space-heating energy for the periods, numerical values of
Reference Year
influencing variables, and normalized space-heating energy. The method
June 50.1
used for normalization is DD
July 105.7
Normal
August 105.4
Space Actual Normal space
September 59.8
Period heating degree-days degree-days heating
Season 321
January 2000 2,497 565 632 2,793
February 2000 2,135 496 593 2,552
Table 5.38 Mean ambient temperatures for Seville
March 2000 1,830 478 547 2,094
April 2000 949 300 384 1,215 Reference Year
May 2000 229 59 148 574 January 10.7
June 2000 66 8 7 58 February 11.9
July 2000 0 0 1 – March 14.0
August 2000 14 0 2 – April 16.0
September 2000 158 153 123 127 May 19.6
October 2000 575 184 305 953 June 23.5
November 2000 1,296 307 443 1,870 July 26.9
December 2000 1,760 448 579 2,275 August 26.8
September 24.4
Sum 11,509 14,512
October 19.5
Minimum – November 14.3
Maximum – December 11.1
96 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

30
26.8 26.8
24.4
25 23.4
Average outdoor temperature (°C)
19.6 19.5
20

16.0
15 14.0 14.3
11.9
10.6 11.1
10

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.14 Monthly average outdoor temperature – Seville

250,000
Global solar radiation
Direct solar radiation

200,000
Solar radiation (Wh/m2)

150,000

100,000

50,000

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.15 Normalized solar radiation in Seville

BUILDING NO. 1 The air-conditioning system is separate for each

General information apartment. It consists of an air-to-air heat pump for both
heating and cooling. The domestic hot-water system is
The building is a six-storey multi-family building constructed an independent boiler without storage. A thermostat in
in 1998. It is located near the city centre of Seville the living room controls the air temperature. Setback
(Figure 5.16). Each apartment has an air-conditioned floor temperatures are selected during both winter and summer
area of 90 m2 and a gross area of some 130 m2 . The seasons.
exterior walls consist of 12 cm of brick, 4 cm of insulation There are individual meters for electricity, natural gas and
(polystyrene), 4 cm of hollow brick and an interior finishing water (but not for hot water). Electricity bills are received
of 1.5 cm of plaster. The wall of the west façade is covered every two months; water and natural gas every three months.
outside by 2 cm of white stone. All the windows are double-
glazed. The ratio of windows to walls is very high in the west
façade. Available data – calculation and results
At present, five people occupy the considered apartment The data are given in Tables 5.39 to 5.44.
throughout the year except for August. The occupancy
pattern is that of a typical Spanish residence. Everybody goes
Remarks
out of the house early in the morning and comes home again
to have lunch at two in the afternoon. Later the occupancy The length of the heating and cooling seasons has been
is more variable, but it could be said that half of the people provided by the owner.
go out and return during the evening. During the weekends, The degree-days for each billed period, both for heating
either Saturday or Sunday is spent away. and cooling, have been calculated from hourly temperature
 Examples and case studies 97

Figure 5.16 Spanish building no. 1: Pictures of the building. The east façade is on the left and the west façade on the right

Table 5.39 Form table 1: Supplied energy There are individual meters for electricity and water (but
Supplied energy Natural gas Electricity not for hot water). Electricity bills are available every two
Estimated (E) or measured (M) M M months; water bills every three months.
Conversion factor 10.8 kWh/Nm3 1.00 kWh/kWh
October–November 1999 810 1,224 Available data – calculation and results
December 1999–January 2000 994 2,280 The data are shown in Tables 5.46 to 5.51.
February–March 2000 940 1,240
April–May 2000 950 1,135 Remarks
June–July 2000 583 1,474
August–September 2000 551 1,149 The lengths of the heating and cooling season have been
Sum (kWh/year) 4,828 8,502 provided by the owner.
The degree-days for each billed period, both for heating
and cooling, have been calculated from hourly temperature
records during the actual periods. The results are shown in records during the actual periods.
Table 5.45. The hot-water consumption has been estimated assuming
45 l/person day, a supply temperature of 45◦ C and a mains
water temperature of 18◦ C.
BUILDING NO. 2
General information
Greek buildings
The building is a two-storey single-family house built in 1977
INTRODUCTION
(Figure 5.17). It is located near the city centre of Seville. The
house has an air-conditioned floor area of 200 m2 and a gross In Greece the energy billing system is mainly concerned with
area of some 240 m2 . The exterior walls consist of 12 cm of electricity and fuel-oil use. Electricity bills and consumption
brick, 4 cm of insulation (polystyrene) and 12 cm of brick. measurements are distributed every three months to all
All the windows are single glazed. The ratio of windows to residences by the Greek Public Power Corporation (PPC).
walls is very high in the south façade, but an overhang is Usually, there is a single meter per household for all
available. electricity uses (hot water, cooking, air conditioning etc.).
At present five people occupy the building throughout the In some cases there may be a second night-rate electricity
whole year except for August. The occupancy pattern is that meter.
of a typical Spanish residential house. Everybody goes out of Diesel oil is the main fuel for central heating systems.
the house early in the morning and comes home again to For a single house the fuel bill comes with the purchase of
have lunch at two in the afternoon. Later the occupancy is fuel (once or twice in a year). For a multi-storey apartment
more variable, but it could be said that half of the people go building the fuel bills are distributed every month by the
and return during the evening. During summer weekends, building manager, together with the other common-use
the house is usually empty. charges (according to the floor area for older flats). In newer
The air-conditioning system is centralized for the whole apartment blocks the occupants of each apartment have the
house. It consists of an air-to-air heat pump for both option to turn on and off the heating depending on their
heating and cooling. The domestic hot-water system consists needs and each flat has its own metering device for billing
of three electrically heated tanks with a storage capacity purposes.
of 80 litres each and a power of 1.5 kW. A thermostat Central heating systems are mostly of the hot water/
in the living room controls the air temperature. Setback radiator type. There is a central boiler room and a fuel tank.
temperatures are selected during both winter and summer Some very new buildings that are going to use natural gas
seasons. are connected directly to the gas-piping network.
98 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.40 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances and lighting Electricity 1.00 1.00 E
Natural gas boiler for hot water Natural gas 0.65 0.65 E
Split-unit heat pump for space heating Electricity COP 3.3 – E
Split-unit heat pump for air conditioning Electricity – COP 2.7 E

Table 5.41 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1
Natural gas boiler Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
October–November 1999 526
December 1999–January 2000 645
February–March 2000 610
April–May 2000 617
June–July 2000 378
August–September 2000 358
Sum 3,134
Method Footnote 2 Footnote 3 Footnote 4
Electrical conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
October–November 1999 177 1,047
December 1999–January 2000 1,251 1,029
February–March 2000 176 1,064
April–May 2000 1,135
June–July 2000 392 1,082
August–September 2000 546 603
Sum 1,604 938 5,960
Total (kWh/year) 1,604 938 5,960 3,134
1. The hot water is supplied by the natural gas boiler unit (at around 55◦ C). Mean yearly efficiency has been estimated as:

Wc kg × 4.18 kJ/kg K × (55 − 18) K/3,600 kJ/kWh/Qsupplied hot water

where Wc = 4 persons × 50 litres/day per person × number of days in the period.

2. Qheating = Qelectricity − Qappliances for each period.
3. Qcooling = Qelectricity − Qappliances for each period.
4. Based on the intermediate periods (neither heating nor cooling) we can estimate a uniform daily consumption by appliances of 17.74 kWh and thus
Qappliances ≈ 17.74× number of days in the period.

Table 5.42 Actual supplied energy for heating and cooling

Actual heating Actual heating Actual cooling Actual cooling
energy consumption energy consumption energy consumption energy consumption
Year (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year)
October 1999–September 2000 1,604 17.82 938 10.42

Table 5.43 Actual delivered energy for heating and cooling

Actual heating energy Actual heating energy Actual cooling energy Actual cooling energy
consumption consumption consumption consumption
Year (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year)
Normalized 5,293 58.81 2,532 28.13

The Athens reference meteorological year has been CSI estimation from a correlation of degree-days and
derived from the records of five official meteorological insolation hours in Greece is calculated as follows:
stations over the last 25 years. Tables 5.52 to 5.54 give some
of the relevant climatic variables. CSI = m1DD + m2 (n/N ) + m3DD 2 + m4n/N 2 + b
 Examples and case studies 99

Table 5.44 Normalized delivered energy for heating and cooling

Normalized heating energy Specific heating energy Normalized cooling energy Specific cooling energy
consumption consumption consumption consumption
Normalization∗ (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year)
Normalized #1 7,786 86.51
Normalized #2 7,818 86.86
Normalized #3 6,886 76.51 1,846 20.51
∗ Three methods have been used in order to obtain the normalized results.

Normalized #1: Normalized heating energy consumption = Actual heating energy consumption × DDnormalized year /DDactual year .
Normalized #2: (Normalized heating energy consumption − Constant × winter solar radiation) = (Actual heating energy consumption − Constant ×
winter solar radiation) × DDnormalized year /DDactual year .
Normalized #3: Normalized heating energy consumption = Actual heating energy consumption × CSInormalized year /CSIactual year and Normalized cooling
energy consumption = Actual cooling energy consumption × CSInormalized year /CSIactual year .

Table 5.45 Heating and cooling degree-days Table 5.46 Form table 1: Supplied energy
Period DDh DDc Supplied energy Electricity
3 October–1 December 1999 167 0 Estimated (E) or measured (M) M
1 December 1999–4 February 2000 431 0 Conversion factor 1.00 kWh/kWh
4 February–5 April 2000 161 0 January–February 1999 3,549
5 April–8 June 2000 7 22 March–April 1999 1,450
8 June–8 August 2000 0 317 May–June 1999 2,391
8 August–3 October 2000 0 174 July–August 1999 2,250
September–October 1999 1,658
where n is the actual number of insolation hours for each November–December 1999 3,580
period, and N is the maximum number of insolation hours Sum (kWh/year) 14,878
for that period.
The coefficients given in Table 5.55 are valid for both
winter and summer periods. The cooling degree-days at 20◦ C separate apartments. Only one apartment (situated on the
are given in Tables 5.56 and a comparison of simple degree- top floor) is considered within this procedure.
days and the results of the CSI method is given in Table 5.57. The total heated area is equal to 78 m2 . Two adults and two
From the above results we see that there is a considerable children occupy the apartment. A natural ventilation system
difference between the DD and CSI normalization ratios, is used (opening windows).
especially during the cooling season. For this reason the A central heating system is installed running on diesel oil.
CSI method was considered to be more suitable for The distribution of heat is realized via a water-borne sys-
Greece. tem and radiators. The apartment is also equipped with a
Figures 5.18 and 5.19 show the data relative to the reversible heat pump running on electricity (split-unit air
reference year. conditioner 2.8 kW, rated power 1.4 kW – coefficient of per-
formance COP ≈ 2.0), which is used for auxiliary heating and
BUILDING NO. 1 cooling, from 2 to 4 hours per day during hot summer days.
Hot-water production (approximately 45% of the total
General information
water consumption) is carried out with the main heating
The building (Figure 5.20) considered is a two-storey system during the winter and with a back-up electrical boiler
detached building constructed in 1992 and comprising two during the summer.

Figure 5.17 Spanish building no. 2: Pictures of the building. The street façade is on the left and the garden façade on the right
100 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.47 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances and lighting Electricity 1.00 1.00 E
Split-unit heat pump for space heating Electricity COP 3.3 – E
Split-unit heat pump for air conditioning Electricity – COP 2.7 E

Table 5.48 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of specific energies
Method Footnote 1 Footnote 2 Footnote 3 Footnote 4
Electrical conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January–February 1999 2,237 761 551
March–April 1999 841 609
May–June 1999 1,033 788 570
July–August 1999 823 828 599
September–October 1999 231 828 599
November–December 1999 2,153 828 599
Sum 4,390 2,087 4,873 3,528
Total (kWh/year) 4,390 2,087 4,873 3,528
1. Qheating = Qelectricity − Qappliances for each period.
2. Qcooling = Qelectricity − Qappliances for each period.
3. Based on the intermediate periods (neither heating nor cooling) we can estimate a uniform daily consumption by appliances of 13.35 kWh and thus
Qappliances ≈ 13.35 × number of days in the period.
4. The hot water is supplied by the electrical heated tanks (at around 55◦ C). Mean yearly efficiency has been estimated as:

Wc (kg) × 4.18 (kJ/kg K) × (55 − 18) (K)/3,600 (kJ/kWh)/Qsupplied hot water

where Wc = 5 persons × 50 litres/day per person × number of days in the period.

Table 5.49 Actual supplied energy for heating and cooling

Actual heating energy Actual heating energy Actual cooling energy Actual cooling energy
consumption consumption consumption consumption
Year (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year)
January–December 1999 4,390 19.95 2,087 9.49

Table 5.50 Actual delivered energy for heating and cooling

Actual heating energy Actual heating energy Actual cooling energy Actual cooling energy
consumption consumption consumption consumption
Year (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year)
January–December 1999 14,487 65.85 5,634.9 25.61

Table 5.51 Normalized delivered energy for heating and cooling

Normalized heating energy Specific heating energy Normalized cooling energy Specific cooling energy
consumption consumption consumption consumption
Normalization* (kWh/year) (kWh/m2 · year) (kWh/year) (kWh/m2 · year)
Normalized #1 21,308 96.86
Normalized #2 21,340 97.00
Normalized #3 18,847 85.67 8,127 36.94
*Three methods have been used in order to obtain the normalized results.
Normalized #1: Normalized heating energy consumption = Actual heating energy consumption × DDnormalized year /DDactual year .
Normalized #2: (Normalized heating energy consumption − Constant × winter solar radiation) = (Actual heating energy consumption − Constant ×
winter solar radiation) × DDnormalized year /DDactual year .
Normalized #3: Normalized heating energy consumption = Actual heating energy consumption ×CSInormalized year /CSIactual year .
 Examples and case studies 101

Table 5.52 Mean ambient temperature Table 5.55 Values of the coefficient to use in the CSI relationship
Year Reference 2000 m4 m3 m2 m1 b
January 10.0 7.4 Winter 1.88 9.04E–06 −2.20 −0.002 0.96
February 8.1 10.5 Summer −5.52 −6.62E–05 8.93 0.028 −4.81
March 10.9 12.3
April 16.6 17.9
May 21.6 24.7 Table 5.56 Cooling degree-days at 20◦ C – normalization ratio
September 23.4 22.3 NRc = 0.739
October 16.6 19.0 Year May June July August September Season
November 15.6 17.3 Reference 73 157 202 230 113 775
December 12.0 12.5 2000 89 211 319 284 145 1,048

Table 5.53 Heating degree-days at 20◦ C for the heating season – Table 5.57 CSI normalization ratio for heating – NRh = 0.994 and for
normalization ratio NRh = 1.146 cooling – NRc = 0.980
Year Reference 2000 Year DDh (n/N )h CSIh DDc (n/N )c CSIc
January 309 389 Reference 296 0.507 0.506 176 0.792 1.698
February 334 276 2000 299 0.527 0.509 240 0.841 1.734
March 281 240
April 116 86
May 23 16 Table 5.58 Form table 1: Supplied energy
September 11 4 Supplied energy Diesel oil Electricity
October 128 53 Estimated (E) or measured (M) M M
November 136 86 Conversion factor 11.92 kWh/kg 1.00 kWh/kWh
December 247 231 November 1999–January 2000 7,212
Season 1,584 1,382 February–April 2000 8,666
November 1999–February 2000 1,281
Table 5.54 Cooling degree-days at 26◦ C for the cooling season – March–June 2000 1,292
normalisation ratio NRc = 0.634 July–October 2000 1,191
Year May June July August September Season Sum (kWh/year) 15,877 3,764

Reference 8 35 58 71 22 194
2000 0 56 131 98 20 306
takes into consideration both the average monthly degree-
days and the solar availability, it is considered to be more
Two electricity meters (one for the cheaper night rate) suitable for countries like Greece, where weather data may
measure the whole consumption of the apartment. The vary substantially from year to year.
bills based on these measurements are available every four
months. Bills for the diesel oil are available at each delivery.
The bills relating to the water consumption are available BUILDING NO. 2
every three months. General information
The building considered is a four-storey semi-detached
Available data – calculation and results building (Figure 5.21) constructed in 1992 and comprising
The information given in Tables 5.58 and 5.60 is available eight apartments, with two on each floor. Only one
from the bills, with the calculated system efficiencies given apartment (situated on the third floor) is considered within
in Table 5.59, based on the hot-water supply temperatures this procedure.
given in Table 5.61. The total heated area is equal to 91 m2 . Two adults and two
Tables 5.62 and 5.63 give the actual and normalized (DD children occupy the apartment. A natural ventilation system
and CSI) values. is used (opening windows).
A central heating is installed, running on diesel oil. The
distribution of heat is realized via a water-borne system and
Remarks
radiators. The apartment is also equipped with a reversible
Approximately 10 to 15% of the boiler energy production heat pump used for auxiliary heating and cooling (split-unit
was used during the heating season for the production of hot air conditioner 2.8 kW rated power 1.4 kW – PER ≈ 2.0)
water. During this period the heating was on for an average running on electricity from 3 to 4 hours per day during hot
of nine hours per day. summer days.
From the above results we see that there is a considerable The hot-water production (approximately 45% of the total
difference between the DD and CSI normalization results, water consumption) occurs via a solar water heater with a
especially during the cooling season. Since the CSI method back-up electrical heater for the winter season.
102 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

30.0
27.4
26.5
25.2
25.0 23.4
Average outdoor temperature (ºC)

21.6

20.0
16.6 16.6
15.6
15.0
12.0
10.9
10.0
10.0
8.1

5.0

0.0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.18 Monthly average outdoor temperature, Athens, Greece – normal year

250,000
Global solar radiation
Direct solar radiation

200,000
Solar radiation (Wh/m2)

150,000

100,000

50,000

0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Figure 5.19 Monthly solar energy available on a horizontal surface, Athens, Greece – reference year

Figure 5.20 Greek building no. 1: Pictures of the building

 Examples and case studies 103

Table 5.59 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E ), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances and lighting Electricity 1.00 1.00 E
Electrical boiler for hot water, 4 kW (1992) Electricity 1.00 E
Central boiler for space and hot-water heating (1992) Diesel oil 0.85 E
Split-unit heat pump for air conditioning Electricity 1.00 COP 2.00 E

Table 5.60 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1 Footnote 2
Central heating system Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
October 1999–January 2000 5,429 701
February–April 2000 6,626 740
Sum 12,058 1,441
Method Footnote 3 Footnote 4 Footnote 5
Electrical conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
November 1999–February 2000 1,281
March–June 2000 168 846 362
July–October 2000 420 437 544
Sum 588 2,564 906
Total (kWh/year) 12,058 588 2,564 2,347
1. Diesel oil quantities delivered for the central heating plant were 605 and 727 kg.
2. During the winter season the hot water is supplied by the central heating economizer unit (at around 60◦ C). i.e.

Wc kg × 4.18 kJ/kg K × (60 − Tw ) K/3,600 kJ/kWh

where Wc = 4 persons × 40 litres/day per person × number of days in the period and Tw is the mean water supply temperature in ◦ C (see Table 5.61).
3. The energy for space cooling can only be estimated by the hours of use of the air-conditioning unit, i.e. QSpCelectric = 2.8 kW × 2 h/day × 30 days =
168 kWh, and QSpCelectric = 2.8 kW × 3 h/day × 50 days = 420 kWh (during this period there is a minimum of three weeks’ vacation).
4. The energy for electrical appliances is estimated as the total delivered electricity minus the electricity for hot water and half of the electricity for
space cooling.
5. During the summer, back-up electricity is used for hot water (minimum 50◦ C) i.e.

Wc kg × 4.18 kJ/kg K × (50 − Tw ) K/3,600 kJ/kWh

(During this period there is a minimum of three weeks vacation.)

Table 5.61 Calculation of the hot tap-water consumption

Available data – calculation and results
Number of days Tw (◦ C) Qhotwater (kWh)
January 31 17 239 The information given in Tables 5.64 and 5.66 is available
February 28 16 245 from the bills, with the calculated system efficiencies given
March 31 15 250 in Table 5.65, based on the hot-water supply temperatures
April 30 16 245 given Table 5.67.
May 31 17 184 Tables 5.68 and 5.69 give the actual and normalized (DD
June 30 18 178 and CSI) delivered and supplied energy consumptions.
July 31 20 167
August 31 21 54
September 30 22 156 Remarks
October 31 20 167 Approximately 65–70% of the annual hot-water energy
November 30 19 228 demand is supplied by solar energy. This amount, which
December 31 18 234 is included in Tables 5.68 and 5.69, is indirectly billed
energy (actually prepaid), the cost of which is based on the
One electricity meter measures the whole consumption investment capital and payback period.
of the apartment. The bills, based on these measurements, During this heating period the system was on for an
are available every four months. Bills for the diesel oil average of eight to nine hours per day.
are provided monthly by the building manager. The bills From the above results we see that there is a considerable
relating to the water consumption are available every 3 difference between the DD and CSI normalization results,
months. especially during the cooling season. Since the CSI method
104 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.62 Actual and normalized – delivered energy consumption

Space heating Area-specific space heating Space cooling Area-specific cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 12,058 154.6 588 7.6 17,557 225.1
DD normal 13,818 177.2 435 5.6 19,164 245.7
CSI normal 11,986 153.7 576 7.4 17,473 224.0

Table 5.63 Actual and normalized – supplied energy consumption

Space heating Area-specific space heating Space cooling Area-specific cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 14,181 181.8 294 3.8 19,641 251.8
DD normal 16,251 208.4 217 2.8 21,633 277.3
CSI normal 14,096 180.7 288 3.7 19,549 250.6

Figure 5.21 Greek building no. 2: Pictures of the building

Table 5.64 Form table 1: Supplied energy energy supplied to the detached house and the apartment
Supplied energy Diesel oil Electricity (Table 5.70) can be attributed mainly to the inclusion of the
Estimated (E) or measured (M) M M solar energy, which was utilized for hot-water heating during
Conversion factor 11.92 kWh/kg 1.00 kWh/kWh the whole year.
November 1999 2,385 A simple calculation of the heat-loss factors was also
December 1999 5,278 performed (Table 5.71), with the mean ambient temperature
January 2000 2,640 and a heating season of five months (3,600 hours) taken into
February 2000 5,191 account for the year 2000. The set-point temperature was
March 2000 2,253 used as the mean indoor temperature for each house. These
November 1999–February 2000 2,198 were reported by the occupants; for the apartment this was
March–June 2000 2,079 set at 22◦ C because of the presence of an infant child.
July–October 2000 2,268 The heat-loss factors turned out to be the same, although
Sum (kWh/year) 17,748 6,545 there is a 15% difference in the floor areas of the two
houses and they are also exposed differently to the weather. A
logical explanation could be that, as the apartment is twice
takes into consideration both the average monthly degree- as high above the ground as the detached house and is several
days and the solar availability, it is considered to be more kilometres away, it experiences different microclimatic
suitable for countries like Greece, where weather data may conditions, which affect the energy performance of the
vary substantially from year to year. building.

COMMENTS ON THE RESULTS FOR GREEK BUILDINGS 1 AND 2 Conclusions concerning the BEP procedure
The results from the two houses in the Athens area are The BEP procedure has been applied to eight buildings in
comparable (when the similar construction, occupancy and four countries. No major problems have been encountered
weather conditions are considered). CSI normalized specific when applying the protocol.
energy supplied for heating is around 186 kW/m2 year, Because of the situations in the different countries,
while the 36 kW/m2 difference in the normal specific global different normalization techniques have been applied within
 Examples and case studies 105

Table 5.65 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances and lighting Electricity 1.00 1.00 E
Electrical/solar hot-water heater, 4 kW (1992) Electricity 1.00 E
Solar 0.40 1.00 E
Central boiler for space heating (1992) Diesel oil 0.85 E
Split-unit heat pump for air conditioning Electricity 1.00 COP 2.00 E

Table 5.66 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1
Central heating system Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
November 1999 2,027
December 1999 4,487
January 2000 2,244
February 2000 4,412
March 2000 1,915
Sum 15,086
Method Footnote 2 Footnote 3 Footnote 4
Electrical conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
December 1999–March 2000 1,617 581
April–July 2000 252 1,806 147
August–November 2000 560 1,717 271
Sum 812 5,141 998
Method Footnote 5
Solar conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
December 1999–March 2000 387
April–July 2000 627
August–November 2000 390
Sum 1,405
Total (kWh/year) 15,086 812 5,141 2,403
1. Diesel oil quantities delivered for the central heating plant were 200.1, 442.8, 221.5, 435.5 and 189 kg.
2. The energy for space cooling can only be estimated by the hours of use of the air conditioning unit, i.e. QSpCelectric = 2.8 kW × 3 h/day × 30 days =
252 kWh, and QSpCelectric = 2.8 kW × 4 h/day × 50 days = 560 kWh (during this period there is a minimum of three weeks vacation).
3. The energy for electrical appliances is estimated as the total delivered electricity minus the electricity for hot water and half of the electricity for
space cooling.
4. During the winter the hot water is partly supplied by the solar heater – at least 40% of the load – with an electric heater as back-up (at around 60◦ C),
i.e. Qhotwater × 40%

Qhot water = Wc kg × 4.18 kJ/kg K × (60 − Tw ) K/3,600 kJ/kWh

where Wc = 4 persons × 40 litres/day per person × number of days in the period and is the Tw mean water supply temperature in ◦ C (see Table 5.67).
5. During the summer season 100% of the load is supplied by the solar heater (minimum temperature around 50◦ C), i.e.

Qhot water = Wc kg × 4.18 kJ/kg K × (50 − Tw ) K/3,600 kJ/kWh

(During this period there is a minimum of three weeks vacation.)

the tests. Depending on the number of energy bills available Estimations of the energy consumption have to be made
during one year, the assessment of the heating/cooling when no bills are available, for instance when wood is
energy consumption requires more or fewer assumptions. burned.
When only one bill per year is available, assumptions about In the BEP procedure, the real indoor climate is not
the energy consumption for hot water or for cooking have taken into account and estimation of the system efficiency
to be made. If energy bills are available on a monthly basis, is performed.
the analysis of the bills when neither heating nor cooling is The systematic application of the BEP procedure can
applied makes it possible to identify the energy consumption easily be realized via the software that has been
for hot water or for cooking. developed.
106 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.67 Calculation of the hot tap-water consumption Table 5.70 Energy supplied to the two Greek buildings
Number of days Tw (◦ C) Qhotwater (kWh) Qheating qH Qglobal qG
January 31 17 239 Building kWh kWh/m2 kWh kWh/m2
February 28 16 245 Detached house 14,096 180.7 19,549 250.6
March 31 15 250 Apartment 17,642 193.9 26,163 287.5
April 30 16 245
May 31 17 184
Table 5.71 Heat-loss factors
June 30 18 178
July 31 20 167 Qdelivered Ts Ta HLF
Building kWh ◦C ◦C W/K
August 31 21 54
September 30 22 156 Detached house 11,986 20 12 416.2
October 31 20 167 Apartment 14,995 22 12 416.5
November 30 19 228
December 31 18 234
average indoor temperature during the monitoring period
was 17.1◦ C. The average outdoor temperature was 2.5◦ C.
RESULTS FOR THE MONITORED ENERGY A summary of the results obtained during the energy
PROTOCOL (MEP) PROCEDURE monitoring is given in Table 5.72.
Introduction
Available data – calculation and results
The MEP protocol has been described extensively in
Chapter 2 and is comparable with the BEP procedure. The Based on the information from the monitoring, the new
following elements are different: supplied energy consumption was calculated, as shown in
Table 5.73.
• Monitoring of the indoor and outdoor climate is carried A comparison between the results obtained in the BEP
out. and MEP procedures is shown in Figure 5.23. The results
• Energy monitoring is carried out. are presented with an error band calculated by applying a
• The efficiency of the heating system is measured for most variation of ±3◦ C to the correction factor for normalization.
of the time. As an extra service within MEP, a normalization of the
indoor temperature is also carried out. Table 5.74 presents
Belgian buildings the results of the calculations performed with this new
information:
BUILDING NO. 1
The monitoring carried out 1. Normalization with reference to the external climate
takes into account the specific values monitored during
Monitoring was carried out for 23 days in this house. The the 23 days of the monitoring period. The normal DD
outdoor and indoor temperatures have been recorded. have been calculated for this period.
Energy monitoring based on manual observation has also 2. Normalization with reference to the indoor climate
been carried out during this period. Figure 5.22 presents assumes that the measured average indoor temperature
the results of the temperature monitoring. (17.1◦ C) is equal to the average indoor temperature
At the beginning of the monitoring, the occupants were during the whole heating season. The degree-days
not present. calculated for this specific temperature are compared
The monitoring period was very cold; during the 23 days with the reference degree-days calculated for a house by
of the monitoring about 13% of the heating degree-days applying a night-time setback (17◦ C average indoor
15/15 of a normal heating season were registered. The temperature).

Table 5.68 Actual and normalized – delivered energy consumption

Space heating Area-specific space heating Space cooling Area-specific cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 15,086 165.8 812 8.9 23,442 257.6
DD normal 17,289 190.0 600 6.6 25,433 279.5
CSI normal 14,995 164.8 796 8.7 23,335 256.4

Table 5.69 Actual and normalized – supplied energy consumption

Space heating Area-specific space heating Space cooling Area-specific cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 17,748 195.0 406 4.5 26,277 288.7
DD normal 20,339 223.5 300 3.3 28,762 316.0
CSI normal 17,642 193.9 398 4.4 26,163 287.5
 Examples and case studies 107

25
Average iindoor
d temperature
No Occupants Outdoor temperaturee
Outd
20

15
Temperature (ºC)

10

-5
05/01 07/01 09/01 11/01 13/01 15/01 17/01 19/01 21/01 23/01 25/01
Date
Figure 5.22 Results of the temperature monitoring

Table 5.72 Data obtained during the energy monitoring (in kWh)
Energy conservation Hot-water Cooking Energy consumption
– Gas consumption consumption Qheat – electricity Qtot
Measured (M), estimated (E) M E E C M C
or calculated (C)
Period Number of days
1 13.28 2,513.0 103.4 35.6 2,374.0 192.5 2,705.4
2 7.51 1,260.9 58.5 20.1 1,182.2 141.4 1,402.3

Table 5.73 Supplied normalized heating and global energy consumption (in kWh/year)
Year Qheat, sup, normal qheat, sup, normal Qtot, sup, normal qtot, sup, normal
Normal energy 27,585 101.0 35,796 131.1
27,172 < Q < 27,585 99.5 < Q < 101.0 35,384 < Q < 35,796 129.6 < Q < 131.1

Table 5.74 Delivered normalized heating energy consumption

during an extremely cold period for Belgium. The lowest
Qheat, del, normal qheat, del, normal
average temperature on a daily basis was equal to −4.2◦ C.
Year (kWh/year) (kWh/m2 year)
10.6% of the normal degree-days have been registered
Normal external 16,551 60.6 during this 14-day period.
Normal indoor 16,026 59 A summary of the results obtained during the energy
monitoring is given in Table 5.75.
BUILDING NO. 2
The monitoring carried out Available data – calculation and results
Monitoring was carried out in the house for 14 days: Based on the information from the monitoring, the new
supplied energy consumption was calculated and is shown
• Manual energy monitoring was performed by the in Table 5.76.
occupants and the different energy consumptions A comparison between the results obtained from the BEP
recorded. and MEP procedures is shown in Figure 5.25. The results
• Monitoring of the indoor and outdoor climates was are presented with an error band calculated by applying a
carried out simultaneously. Four temperatures inside the variation of ±3◦ C to the correction factor for normalization.
house were recorded as well as the outdoor As an extra service within MEP, a normalization of the
temperature. indoor temperature is also carried out. Table 5.77 presents
The results of the temperature monitoring are shown in the results of the calculations performed with this new
Figure 5.24. information:
The average daily indoor temperature measured was
16.0◦ C. The average outdoor temperature during the 1. The normalization with reference to the external
monitoring was 0.3◦ C. The monitoring was carried out climate takes into account the specific values monitored
108 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

200
qAsup, glob, N
180
Specific normalized energy consumption (kWh/year)
164 168 qAsup, heat, N
161
160

140 131
125 120
119 118
120
101
100 89
80

60

40

20

0
BEP - 1996-1997 BEP - 1997-1998 BEP - 1998-1999 BEP - 1999-2000 MEP - normal
Figure 5.23 Comparison of the energy consumption

20

15
Daily average temperature (ºC)

10

0
Indoor
Outdoor
-5
08/01 10/01 12/01 14/01 16/01 18/01 20/01
Date
Figure 5.24 Results of the indoor climate monitoring

Table 5.75 Data obtained during the energy monitoring (in kWh)
Energy consumption Hot-water Cooking Energy consumption
– gas consumption consumption Qheat – electricity Qtot
Measured (M), estimated (E) M M E C M C
or calculated (C)
Period Number of days
1 6.55 643.7 50.9 25.7 567.1 63.7 712
2 7.14 767.4 51.4 28.1 688.0 125.1 903

Table 5.76 Supplied normalized heating and global energy consumption (in kWh/year)
Year Qheat, sup, normal qheat, sup, normal Qtot, sup, normal qtot, sup, normal
Normal external 11,795 42.7 21,020 76.1
11,795 < Q < 13,906 42.7 < Q < 50.4 21,020 < Q < 23,130 76.1 < Q < 83.6
 Examples and case studies 109

Figure 5.25 Comparison of the energy consumption

Table 5.77 Delivered normalized heating energy consumption values from overall values, assess what was supplied as specific
Qheat, del, normal qheat, del, normal energy for appliances and lighting.
Year (kWh/year) (kWh/m2 year) The electric boiler is utilized for the purpose of space
Normal external 15,765 57.1 heating (10.5 kW) and for heating of tap water (9 kW).
Normal internal 18,817 68.1 Measurements were conducted on the coils for tap-water
heating, in other words, the energy delivered to the water
was measured. Tap water is heated with three coils, each
during the 14-day monitoring period. The normal DD having a power of 3 kW. With the help of run-time sensors on
have been calculated for this period. each coil, the delivered energy could be assessed by multi-
2. The normalization with reference to the indoor climate plying by the power of each coil. As for the space-heating
assumes that the measured average indoor temperature coils, measurement of the current supplied to the unit has
(16.0◦ C) is equal to the average indoor temperature been realized directly on the fuse.
during the whole heating season. The degree-days
calculated for this specific temperature are compared
Wood measurements
with the reference degree-days calculated for a house by
applying a night-time setback (17◦ C average indoor The amount of wood used for space-heating purposes was
temperature). weighed and summed on a daily basis. The occupants
performed the weighing.

Swedish buildings
Temperature measurements
BUILDING NO. 1
Two thermal zones were predicted to exist on the basis that
The monitoring carried out there were two storeys. The fan stove is situated on the lower
Monitoring of the building was primarily concerned with storey next to an open staircase. Since the stove primarily
measurement of the specific energies supplied by electricity blows warm air, it was considered that the upper storey would
and wood for the fan stove and with measurement of be warmer than the lower storey.
temperatures. Measurements were conducted with thermistor sensors
and a four-channel data logger. Two sensors were used for
the thermal zones, a third for measurement of the external
Electricity environment, while the fourth was placed in the air outlet of
The electricity suppliers make measurements on the overall the fan stove.
electricity supplied. Hourly values are registered and these
are collected on a daily basis by means of the telephone;
Available data – calculation and results
at midnight the company automatically dials the customer
number and data is silently transferred. These values can be Measurements were initiated on 17 January (day number
seen on the bill. 17) and terminated on 19 February (day number 50). A
With the overall electricity measured by the supplier, the part of the series is shown in Figure 5.26. A temperature
task was to measure how this electricity was distributed within measurement was performed on the fan stove warm air
the residence. The strategy was to measure electricity that outlet – a type of control for inspection of when it was
was supplied to the boiler units and, by subtracting these in use. Contrary to prediction, the differences in air
 110 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Max 50.3 Max 21.3 Max 21.5 Max 9.4

Avg 28.98 Avg 19.43 Avg 19.6 Avg 0.7521
Min 20.1 Min 18.8 Min 18.3 Min -4.3
= T1 = T2 = T3 = T4
22.3 19.3 19.6 0.8
50

35

20

-10
20010216 20010216 20010217 20010217 20010218 20010218 20010219 20010219
020000 140000 020000 140000 020000 140000 020000 140000
20010218
170309
Figure 5.26 A sequence of four measured temperatures. T1 measures warm air from the fan stove. T2 is the temperature of the lower storey and T3 that
of the upper storey. T4 is the external temperature

140 40

120 35

30
100
Wood energy [kWh/day]

25
Energy [kWh]

80
20
60
15

40
10

20
5

0 0
15 20 25 30 35 40 45 50 10 15 20 25 30 35 40
Day number Temperature difference [K]
Figure 5.27 Monitored energy use (day number 17 = 17 January and day Figure 5.28 Wood use per day as a function of temperature difference.
number 50 = 19 February). The dotted line represents daily energy for tap- The spread of points does not show a clear correlation between use of
water heating. The filled line corresponds to daily use of space heating and wood and the temperature difference
appliances. The dashed line is the sum of the latter energy use including
the space heating delivered by the fan stove

Analysis of wood consumption

temperatures between the storeys are surprisingly small. The occupants were requested to weigh the wood that was
One reason is the mechanical ventilation system, which placed in the stove. Daily values were noted.
counteracts stack effects. For wood consumption, two calculation schemes are
The occupants were away for four weekends, which can shown. The first is made by assuming that the consumption
be seen in the tap-water heating in Figure 5.27. The rise of wood is climate dependent, and the second is based on
afterwards is explained by laundry. The peak in space heating the assumption that wood consumption is evenly distributed
in the middle of the sequence is due to an extreme drop in over the heating season.
temperature. As shown in Figure 5.28, the correlation between wood
The recordings of the current to the space heater and to use and temperature difference is poor. The equation of the
the tap-water heating unit were unfortunately lost. As a result regression line is (0.42 × Temp_diff + 3.1) kWh/day, which
of the intensive sampling frequency (1 second intervals), for the actual year gives a supplied wood heat corresponding
the batteries of the logger ran low and failed to maintain to 2,843 kWh. However, this value includes wood burning
voltage sometime between the end of the monitoring and during the summer, which is not practised. The value for the
the attempted transfer of data to the computer. heating season will only be 2,427 kWh.
 Examples and case studies 111

On the assumption that wood consumption is climate- Table 5.78 Form table 1: Supplied energy
independent, the daily mean corresponds to 11.9 kWh/day. Supplied energy Electricity Wood
With 243 days of heating (the default Swedish heating season Estimated (E), measured (M) M M, C
length), the total is 2,880 kWh. This value may be somewhat or calculated (C)
high, when it is considered that the average daily value is Conversion factor 1.00 kWh/kWh 1,500 kWh/m3
derived from a period when the external temperature was December 1999 1,874
on average −1.9◦ C. January 2000 1,847
Irrespective of which of the two methods was used, the February 2000 1,459
results are in this case fairly consistent. The values estimated March 2000 1,426
from the measurements do not seem to be unreasonable April 2000 1,040
in comparison with the estimation by the occupant of the May 2000 688
annual consumption (some 3,000 kWh). June 2000 509
July 2000 474
Analysis of energy for tap-water heating August 2000 438
September 2000 640
Tap-water heat was measured and recorded on a daily October 2000 769
basis. With the assumption that this entity is constant November 2000 963
throughout the heating season, a straightforward average Heating season December – 2,427
can be calculated on the basis of measured data. The 1999–November 2000
mean value is 7.6 kWh/day, giving an annual value of Sum (kWh/year) 12,127 2,427
2,782 kWh/year. This value may be overestimated for the
following reasons:
prescribed on a national level). The values for delivered
and supplied energy are given in Tables 5.82 and 5.83.
• The incoming water temperature is higher during the
summer period.
The heat-loss factor
• Shower water temperatures are lower during the summer
period. The extended period may be used for estimation of the heat-
• The residents are away for six weeks during the summer loss factor. The 34 days of monitoring is short, when seen
period. from the point of view of energy measurements. However,
the period was relatively cold and large variations in the
The last point is to some extent taken into consideration external temperature were recorded. The collected data was
in the average value, since the occupants were away during run through a static Energy Signature (ES) model. The
part of the monitoring period. effects of solar radiation and metabolic heat were neglected.
As mentioned in the discussion of BEP, the Swedish The heat delivered by the radiators, appliances and lights,
default value for one person is 1,000 kWh/person per year. and combusted wood was regressed against temperature
The estimation above is higher for the two occupants. differences in the external and internal environments, on
After an inquiry, it emerged that tap hot water is also a daily basis. The results are plotted in Figure 5.29.
supplied to the dishwasher, which leads to an increased The points have a large scatter, mainly due to dynamic
value. effects that would suggest the use of a dynamic model.
Estimates were made for the summer hot-water con- The heat-loss factor, which corresponds to the slope of the
sumption. The period consists of 122 days, during which regressed line, is found to be 140 W/K.
the occupants are away for six weeks. During the six The next method to be applied is a static ES method that
weeks, the house is visited once a week. This gives makes use of bills. With the assumption that the internal
74 days at the house. Furthermore, a modification was temperature is the same throughout the seasons as during
made for the incoming water. Whereas the temperature the monitoring period, monthly energy use can be plotted
is about 6◦ C during winter, the summer temperature is against temperature difference. The results obtained are
about 12◦ C. The set-point temperature is 80◦ C. Energy use plotted in Figure 5.30.
for the summer period is estimated to be 523 kWh (74 The dashed line is regressed through the open circles.
days), and for the heating season 1,852 kWh (243 days). Here, global energy, excluding energy for hot water, is used.
The annual energy for tap-water heating is estimated to The heat-loss factor is 113 W/K on the basis of all the points.
be 2,375 kWh/year. Note the shift in the four left-most points, which depict
Tables 5.78 to 5.81 give details of the measured values, energy use during the summer: these can be excluded from
the system efficiencies and the non-normalized delivered the regression. Exclusion of these gives a heat-loss factor of
energy. 132 W/K. If only the last three points are omitted, the value
In MEP, it is necessary that a calculation is done in which becomes 124 W/K.
the indoor conditions are normalized, especially the indoor The line with open diamonds is a regression with specific
temperature. For this purpose, in accordance with using energy for space heating, as shown in the previous tables.
MUF as the normalization method, the standard set-point These points indicate a heat-loss factor corresponding to
temperature was chosen to be 21◦ C (this type of value is 134 W/K.
112 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.79 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Estimated (E), measured (M)
Energy conversion system Fuel type Winter efficiency Summer efficiency or calculated (C)
Household appliances and lighting Electricity 1.00 1.00 E, C
Electrical boiler for space heating and hot water, 10.5 kW (1982) Electricity 1.00 1.00 M, C
Fan stove, wood manual feeding (installed 1993) Wood 0.85 0.85 E (manufacturer info)

Table 5.80 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 3 Footnote 2 Footnote 1
Electrical conversion systems Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
December 1999 1,243 394 236 0
January 2000 1,216 394 236 0
February 2000 900 356 213 0
March 2000 796 394 236 0
April 2000 430 382 229 0
May 2000 110 394 184 0
June 2000 0 382 128 0
July 2000 0 342 132 0
August 2000 0 306 132 0
September 2000 80 382 178 0
October 2000 139 394 236 0
November 2000 353 382 229 0
Sum 5,256 4,502 2,369 0
Method Footnote 4
Wood fan stove Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
December 1999 309
January 2000 304
February 2000 264
March 2000 273
April 2000 210
May 2000 79
June 2000 0
July 2000 0
August 2000 0
September 2000 81
October 2000 175
November 2000 209
Sum 1,904
Total (kWh/year) 7,160 0 4,502 2,369 0
1. Electricity used to heat the tap water was measured. The daily mean value for the heating season was determined to be 7.6 kWh/day. For the summer
period (15 May to 15 September), the mean value was modified, since the occupants were present some 78 days out of the 122 possible, and a correction
term, (80 − 12)/(80 − 6), was applied to allow for incoming water temperature rising from 6 to 12◦ C.
2. The data for assessing the energy used by appliances was lost. Relying on bills, the month of June was used to assess the mean appliance power by
using the electricity bill value (global) and subtracting the energy for heating water. The difference was divided by month hours, yielding 530 W. For July
and August, the appliance energy is the billed value reduced by the hot-water energy.
3. The space-heating energy delivered from the radiator system is estimated to be the total delivered electricity minus the electricity for tap-water
heating (footnote 1) and the electricity for the appliances (footnote 2).
4. From measurements, the wood energy content was regressed against the temperature difference between the internal and external environments.
The regression line was used, with mean monthly temperatures, to estimate monthly consumption. For the months May and September, the value was
halved because of the definition of the heating season. The total amount of wood was estimated by the occupant to be 2 m3. MEP measurements and
calculations estimated 1.6 m3.

The solid line with regression on the stars is based method is 140 W/K. The fact that this value is exactly
on the heat losses at the set-point temperature (the the same as from the monitored period is probably
EN 832 type of calculation). In other words, only the coincidental.
useful heat gains and space-heating energy are plotted From this analysis, the heat-loss factor lies in the range
for the entire year. The heat-loss factor with this between 130 and 140 W/K.
 Examples and case studies 113

Table 5.81 Form table 4: Display of chosen processing periods, actual delivered space-heating energy for the periods, numerical values of influencing
variables, and normalized space-heating energy. The lowest two rows indicate the band (minimum and maximum) values that are obtained with
prescribed conditions from the normalization methods. The method used for normalisation was MUF
Space Actual Actual Actual Normal Normal Normal Normal
Period heating exterior temp direct solar diffuse solar exterior temp direct solar diffuse solar space heating
December 1999 1,552 −1.6 16,484 4,860 −1.9 13,600 4,500 1,585
January 2000 1,521 −1.2 33,285 7,661 −3.7 18,300 7,400 1,786
February 2000 1,534 −0.1 48,482 16,842 −3.8 38,000 16,400 1,479
March 2000 1,069 1.6 138,989 31,982 −0.5 68,400 34,400 1,320
April 2000 640 6.7 96,270 51,972 4.2 103,700 54,500 863
May 2000 188 12.2 150,266 75,099 10.4 169,000 68,500 320
June 2000 0 14.3 137,121 87,221 15.1 146,800 76,000 5
July 2000 0 16.3 102,765 69,320 16.7 156,700 72,500 4
August 2000 0 15.9 131,930∗ 34,670∗ 15.7 121,800 61,400 8
September 2000 162 11.1 150,235 36,016 11.3 92,700 37,400 169
October 2000 313 10.5 18,314 20,890 7.0 49,800 21,300 571
November 2000 562 6.8 3,286 5,965 1.8 21,300 8,400 938
Sum 7,160 9,050
Minimum 8,919
Maximum 9,827
*Not available; reference year 1971 data used for August.

Table 5.82 Delivered energy. Normalized values were obtained with MUF
Area-specific Area-specific Area-specific
Space heating space heating Space cooling space cooling Global global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
December 1999–November 2000 7,159 60.2 0 0 14,030 117.9
Normal external 9,050 76.0 0 0 15,921 133.8
Normal internal 10,696 89.9 0 0 17,567 147.6

Table 5.83 Supplied energy. Normalized values were obtained with MUF
Area-specific Area-specific Area-specific
Space heating space heating Space cooling space cooling Global global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
December 1999–November 2000 7,534 63.3 0 0 14,366 120.7
Normal external 9,475 79.6 0 0 16,346 137.4
Normal internal 11,197 94.1 0 0 18,068 151.8

5.5
Remarks
5
The MEP results differ from those from BEP. The largest
effect is due to estimations in the non-billed wood consump- 4.5

tion. In BEP, where the occupant supplied the information, 4

wood consumption seems to be overestimated. On the
3.5
Power [kW]

other hand, the estimates in MEP may be erroneous since

there is no clear information on whether the use of wood 3

is dependent on or independent of climate. What MEP 2.5

does is estimate the order of magnitude, providing a type of
2
verification as to whether or not the information supplied is
correct. 1.5
Though the extended period can be used to determine the 1
heat-loss factor of the building, there may be circumstances
where the frequency and quality of the bills are such that, in 0.5
10 15 20 25 30 35 40
combination with short-term MEP, they allow evaluation of Temperature difference [K]
this value. This has been shown for both Swedish buildings. Figure 5.29 Static model applied to daily mean values. The power
It has been possible because the billing frequency is monthly delivered by heating units and appliances is plotted against temperature
and temperatures have been assessed. difference. The effect of solar radiation is neglected
 114 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

4500 23

4000
22

3500
21
3000
Rate of the heat loss [W]

Temperatures [°C]
20
2500

2000 19

1500
18
1000

17
500

0 16
0 2 4 6 8 10 12 14 16 18 20 22 0 5 10 15 20 25
Temperature difference [K] Time of day [h]

Figure 5.30 Regression lines for energy use against temperature Figure 5.31 The dashed lines show the set-point temperature and room
difference. The diamonds are data for delivered specific space heating. temperature for the thermostat in the south-facing room (see plan drawing
The open circles represent global delivered energy excluding hot-water in Figure 5.32). The filled lines are for corresponding temperatures in a
energy. Stars depict calculated heat losses at the set-point temperature north-facing room

Available data – calculation and results

BUILDING NO. 2
The results of measurement and calculation are given in
This building is fully equipped with monitoring equipment Tables 5.84 to 5.87, with the actual and normalized values
for the purpose of measuring specific energy. The system can of the energy in Tables 5.88 and 5.89. The energy values are
measure the temperature at the location of the thermostat not affected by temperature.
controls, but this facility had not previously been applied.
Sampling was carried out every 2 seconds.
Specific energy is continuously measured and billed. The
Remarks
only information that the standard procedures of MEP will For all intents and purposes, the quality of billed data for
add to that provided by BEP relates to temperature. This a building with this type of billing/monitoring system is
means that processed values of BEP do not change. The such that the standard procedures of MEP (which exclude
reasons are as follows: the extra services) are superfluous. The extra information
provided in this case was the internal temperatures. The
• Actual specific energy use does not change; the values of system was capable of recording this data, but had not been
this have been measured and billed for over a year. used in this way prior to the MEP measurements.
• Normalization calculations with MUF make use of The results will therefore be the same as for the
set-point temperatures, not actual temperatures. BEP procedure. The heat-loss factor, based on measured
• The heat-loss factor is not affected by a systematic change (billed) monthly values, will not be influenced by the mean
in internal temperature. temperature since the slope of the line (delivered heat
plotted against temperature difference) is unaffected.
In general, the energy analysis for individual modern
The monitoring carried out Swedish multi-family apartments may tend to become
The apartment is equipped with two thermostat controls, one complicated. With envelope elements that are well insulated,
in the southern thermal zone and the other in the northern transmission may be quite small, which leads to a
zone. These were now triggered to register values. Intensive larger influence of heat exchange between apartments,
measurements were performed for a period of two weeks, via both transmission and air leakage. The temperatures
with values saved every 11.25 minutes. The temperatures for of neighbouring apartments and the characteristics of
3 April 2001 are shown in Figure 5.31. the internal construction may have to be taken into
The purpose of the measurements is to verify that the consideration in the analysis in order to obtain ‘actual’ values
set-point temperatures are correct. Note that the daytime of energy use, especially for space heating.
set-point is higher than in the BEP analysis. The reason
for this is that, during the audit, the occupants noted that Spanish buildings
the apartment was chilly during cold winter periods. The BUILDING NO. 1
auditor showed the occupants how the set-point temperature
The monitoring carried out
could be reset for daytime. The set-point temperature was
raised by 1◦ C to 21.5◦ C. Since the MEP measurements, A period of 40 days, from 27 January to 8 March, was selected
the occupants have not changed the set-point temperatures as representative of the monitoring carried out in this house.
again. During this period the outdoor and indoor temperatures, the
 Examples and case studies 115

Figure 5.32 Plan drawing of the apartment; the downwards direction is south. The thermostat controls, where set-point
temperatures are actuated and room temperatures collected, are indicated by dots. The zone within the shaded boundary is
the corridor and elevator, i.e. common premises

solar radiation and the electricity consumption have been energy consumption with the indoor–outdoor temperature
registered. The electricity consumption was base on manual difference and the solar radiation.
observation.
Figure 5.33 shows the average indoor and outdoor
Available data – calculation and results
temperatures in this period. From this figure it can be
seen how short periods of three days have been grouped A summary of the results obtained during the energy
together in order to obtain results and to correlate monitoring is given in Table 5.90.
116 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.84 Form table 1: Supplied energy

Supplied energy Electricity (appliances) District heating 1 District heating 2 District + electricity
Estimated (E) or measured (M) M M M M
Conversion factor 1.00 kWh/kWh 1.00 kWh/kWh 1.00 kWh/kWh 1.00 kWh/kWh
January 2000 573 2,497 197 656
February 2000 503 2,135 169 589
March 2000 482 1,830 186 582
April 2000 410 949 137 463
May 2000 406 229 202 152
June 2000 369 66 175 230
July 2000 302 0 164 111
August 2000 300 14 164 171
September 2000 329 158 148 208
October 2000 423 575 180 319
November 2000 399 1,296 142 356
December 2000 490 1,760 208 447
Sum (kWh/year) 4,986 11,509 2,072 4,284

Table 5.85 Form table 2: List of energy conversion systems, their efficiencies and how these were assessed
Winter Summer Estimated (E), measured (M)
Energy conversion system Fuel type efficiency efficiency or calculated (C)
Household appliances and lighting Electricity (appliances) 1.00 1.00 E
District heating 1 + 2; 1 = space heating, 2 = hot water District heating 1.00 1.00 E
District heating and electricity of external spaces and appliances District heating + electricity 1.00 1.00 E

These values have been used to obtain the heat-loss factor BUILDING NO. 1
UA in W/K and the solar aperture gA value in m2 , through The monitoring carried out
a linear correlation for which R 2 = 0.81. Therefore, the
energy consumption can be estimated using the following The indoor temperature and humidity sensors were
correlation: placed in the living room and in one of the bedrooms
(see Figure 5.35). The outdoor temperature sensor was
Heating energy consumption = UA(Tin − Tout ) + gA Rad located at a sheltered position on the balcony. The sensors
where for this building UA = 227 W/K, gA = −1.45 m2* . (Tinytags) were installed on 31 January and were removed
Figure 5.34 shows the correlation between the results on 28 February 2001; data were recorded every 30 minutes.
and the measured data. The detailed monitoring of energy use in the house was
The delivered energy for heating can be estimated carried out from 10 to 23 February 2001. The temperature
by taking into account that the temperature difference profiles are shown in Figure 5.36. During this period the
(Tin − Tout ) is equal to the heating degree-days for the average ambient temperature was 11.3◦ C and the mean
winter season. The number of heating degree-days for this indoor temperature 19.9◦ C.
period is 417. At the same time, the total radiation over a
horizontal surface is 395 kWh/m2 if the total radiation for Available data – calculation and results
those days with an average temperature less than 15◦ C is
From the monitoring data the uncertainty in the set-point
added. Thus, the heating energy consumption is 1,698 kWh
control and in the indoor air temperature measurements
(18.87 kWh/m2 ).
may vary between ±1 and 3◦ C. For this reason normalization
ratios were estimated for an indoor temperature of 20◦ C
with an error band of ±2◦ C. Degree-days were calculated
Greek buildings
from the February measurements. Values are given in
INTRODUCTION Tables 5.91 to 5.94.
During the two-week monitoring period the fuel and energy Normalization with respect to indoor climate was done
consumptions were recorded, together with the indoor and for an air temperature of 19.9◦ C – the average during the
outdoor temperatures and the indoor humidity, as well as monitoring period.
the occupants’ habits.
Remarks
Physically, g.A value should always be greater than 0. Looking at the profile of the indoor air temperatures, we
* Thenegative g.A value is due to the mathematical analysis applied see that this follows the pattern of the ambient air, with
on the input data where an inaccuracy exists. a relatively high amplitude. This is because of the low
 Examples and case studies 117

Table 5.86 Form table 3: Allocation of delivered energy, on an energy conversion system level, to each of the specific energies
Method Footnote 1
Household appliances and lighting Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January 2000 573
February 2000 503
March 2000 482
April 2000 410
May 2000 406
June 2000 369
July 2000 302
August 2000 300
September 2000 329
October 2000 423
November 2000 399
December 2000 490
Sum 4,986
Method Footnote 2 Footnote 3
District heating 1 + 2 Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January 2000 2,497 197
February 2000 2,135 169
March 2000 1,830 186
April 2000 949 137
May 2000 229 202
June 2000 66 175
July 2000 0 164
August 2000 14 164
September 2000 158 148
October 2000 575 180
November 2000 1,296 142
December 2000 1,760 208
Sum 11,509 2,072
Method Footnote 4
District heating + Electricity Space heating Space cooling Appliances and lighting Tap hot water External appliances/spaces
January 2000 656
February 2000 589
March 2000 582
April 2000 463
May 2000 152
June 2000 230
July 2000 111
August 2000 171
September 2000 208
October 2000 319
November 2000 356
December 2000 447
Sum 4,284
Total (kWh/year) 11,509 0 4,986 2,072 4,284
1. Electricity delivered to the apartment is measured and billed at the end of each month.
2. Delivered space heating for the apartment is measured and billed at the end of each month.
3. The quantity of tap water, hot and cold separately, is measured. The system operator calculates a temperature increase for hot water corresponding
to 55◦ C times the specific heat capacity of water. This quantity is billed.
4. External energy use (measured) is for common-space heating, lighting, laundries and operation of installations (fans, pumps and elevators).
118 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Table 5.87 Form table 4: Display of chosen processing periods, actual space-heating energy for the periods, numerical values of influencing variables,
and normalized space-heating energy. The bottom two rows indicate the band (minimum and maximum) values that are obtained with conditions
prescribed according to the MUF normalization method
Space Actual external Actual Actual Normal Normal Normal Normal
Period heating temperature direct solar diffuse solar external temp direct solar diffuse solar space heating
January 2000 2,497 −1.2 33,285 7,661 −3.7 18,300 7,400 2,940
February 2000 2,135 −0.1 48,482 16,842 −3.8 38,000 16,400 2,722
March 2000 1,830 1.6 138,989 31,982 −0.5 68,400 34,400 2,299
April 2000 949 6.7 96,270 51,972 4.2 103,700 54,500 1,206
May 2000 229 12.2 150,266 75,099 10.4 169,000 68,500 325
June 2000 66 14.3 137,121 87,221 15.1 146,800 76,000 361
July 2000 0 16.3 102,765 69,320 16.7 156,700 72,500 41
August 2000 14 15.9 131,930∗ 34,670∗ 15.7 121,800 61,400 140
September 2000 158 11.1 150,235 36,016 11.3 92,700 37,400 183
October 2000 575 10.5 18,314 20,890 7.0 49,800 21,300 926
November 2000 1,296 6.8 3,286 5,965 1.8 21,300 8,400 1,931
December 2000 1,760 −1.6 16,484 4,860 −1.9 13,600 4,500 2,346
Sum 11,509 14,932
Minimum 14,200
Maximum 16,067

Table 5.88 Delivered energy. Normalized values were obtained with MUF
Space heating Area-specific space heating Space cooling Area-specific space cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 11,509 108.6 0 0 18,567 175.2
Normal 14,932 140.9 0 0 21,990 207.5

Table 5.89 Supplied energy. Normalized values were obtained with MUF
Space heating Area-specific space heating Space cooling Area-specific space cooling Global Area-specific global
Year kWh/year kWh/m2 · year kWh/year kWh/m2 · year kWh/year kWh/m2 · year
2000 11,509 108.6 0 0 22,851 215.6
Normal 14,932 140.9 0 0 26,274 247.9

thermal mass of the building and the intermittent heating on 2 March 2001; data were recorded every 30 minutes.
and because the daily morning ventilation is accomplished The detailed monitoring of energy use in the house was
by opening the windows. During the heating operation carried out from 15 to 28 February 2001. During this period
the room temperature was always above the set-point the average ambient temperature was 12.3◦ C and the mean
temperature (21 to 23◦ C). indoor temperature 20.6◦ C. The temperature profiles are
Approximately 13% of the boiler energy production was shown in Figure 5.38.
used for the production of hot water during this two-week
period. During this period the heating was on for 126 hours
(9 hours per day). Available data – calculation and results
Since the CSI method takes into consideration both the
From the monitoring data the normalization ratios were esti-
average monthly degree-days and the solar availability, it was
mated for an indoor temperature of 20◦ C with an error band
considered to be more suitable for normalization in Greece.
of ±2◦ C. Degree-days were calculated from the February
measurements. Values are given in Tables 5.95 to 5.98.
BUILDING NO. 2 From the monitoring data the uncertainty in the set-point
control and in the indoor air temperature measurements
The monitoring carried out
may vary between ±1 and 3◦ C. For this reason normalization
The indoor temperature and humidity sensors were ratios were estimated for an indoor temperature of
placed in the living room and in one of the bedrooms 20◦ C with an error band of ±2◦ C.
(see Figure 5.37). The outdoor temperature sensor was Normalization with respect to indoor climate was done
located at a sheltered position on the balcony. The sensors for an air temperature of 20.6◦ C – the average during the
(Tinytags) were installed on 27 January and were removed monitoring period.
 Examples and case studies 119

Indoor Temperature
24 Average Outdoor Temperature

22

20
Air Temperature (°C)

18 No
representative
period

16

14

12

10
27-Jan 30-Jan 02-Feb 05-Feb 08-Feb 11-Feb 14-Feb 17-Feb 20-Feb 23-Feb 26-Feb 01-Mar 04-Mar 07-Mar 10-Mar
Date
Figure 5.33 Results of the temperature monitoring

Table 5.90 Data obtained during the energy monitoring

Energy Energy consumption Solar
consumption (W) – appliances (W) Qheat (W) Tin − Tout (K) radiation (W)
Measured (M), estimated (E) M E C M M
or calculated (C)
Period Number of days
1 3 1,936 194 1,742 8.62 111
2 3 2,394 194 2,200 10.78 140
3 3 1,692 194 1,497 7.87 133
4 3 1,631 194 1,436 7.37 66
5 3 2,349 194 2,154 8.83 133
6 3 974 194 779 4.94 179
7 3 2,074 194 1,879 9.37 146
8 3 1,936 194 1,742 9.09 166
9 5 1,597 194 1,403 7.75 135
10 3 1,524 194 1,329 6.11 71
11 3 1,753 194 1,558 6.13 116

Remarks During this period the heating was on for 83 hours (about 6
hours per day).
Looking at the profile of the indoor air temperatures, Since the CSI method takes into consideration both the
we see that this follows the pattern of the ambient air, average monthly degree-days and the solar availability, it
with a relatively high amplitude. This is because of the was considered to be more suitable for normalization in
low thermal mass of the building and the intermittent Greece.
heating and because the daily morning ventilation was
accomplished by opening the windows. During the heating
COMMENTS ON BEP AND MEP RESULTS FOR THE TWO GREEK
operation the room temperature was always above the set-
BUILDINGS
point tempertature (22 to 24◦ C).
Approximately 40% of the hot-water energy demand The MEP results from the two houses in the Athens area
was supplied by solar energy during this two-week period. are comparable to the BEP results. For the detached house
 120 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

2,500 Table 5.92 Normalization ratios for heating – NR @ 22 = 1.610, NR @ 20 =

1.575 and NR @ 18 = 1.466
Calculated Heating Consumption (W)

Period DD @ 22 DD @ 20 DD @ 18 (n/N) CSI @ 20

2,000
Reference year 277 297 390 0.507 0.505
February 2001 172 232 266 0.583 0.293

1,500
daily in 2000) because of the presence of a new-born baby
(while during the monitoring period in February 2001 this
1,000 was reduced by 30% to 6 hours daily), thus leading to a
lower fuel consumption and energy results. The results are
summarized in Table 5.99.
500 A simple calculation of the heat-loss factors was also
performed with the mean ambient temperature and the
heating delivered during the 14-day period (336 hours)
0 taken into account (Table 5.100). The average recorded
0 500 1,000 1,500 2,000 2,500 temperature was used for the mean indoor temperature of
Measured Heating Consumption (W) each house, although the set-point temperature was 20◦ C in
Figure 5.34 Correlation of the results with the measured data both cases.
From the average outdoor temperatures, the effect of
microclimate is obvious, since the two houses in Athens,
which are only a few kilometres apart, are exposed to
Balcony
different weather conditions. The heat-loss factor values
found above are close to the 416 W/K (from BEP) and
Up the difference can be attributed to the fact that the
Bedroom detached house is more exposed being the top floor (up
Living-Room a hill with a lot of trees all around), when compared to
the apartment, which is on an intermediate floor (in a
heavily built area), although the floor area of the latter is
Bathroom 15% higher.

Entrance
Conclusions concerning the MEP procedure –
Bedroom Kitchen comparison with BEP
The major difference in the MEP procedure when com-
Balcony pared to the BEP procedure is that the accuracy of the final
results must be better.
Datached House - Upper Floor The application of the MEP procedure requires an
energy monitoring to be performed. Several type of energy
Figure 5.35 Plan of the detached house; the downward direction is monitoring have been applied:
south-west
• monitoring of the temperature (indoor and outdoor
temperature)
the CSI normalized specific energy supplied for heating • energy monitoring, carried out on the basis of automatic
is around 190 kWh/m2 year, while the normalized global recording of the consumption or on the basis of manual
energy supplied is 265 kWh/m2 year. readings of the gauges.
The difference between the MEP (2001) and BEP (2000)
results for the apartment house, can be attributed mainly The MEP procedure delivers results for the normalized
to the higher use of the heating system (around 9 hours energy consumption based on the outside climate, as well

Table 5.91 Data obtained during the energy monitoring (in kWh)
Energy consumption Hot water Cooking Energy consumption
– oil consumption consumption Qheat – electricity Qtot
Measured (M), estimated (E) M E E C M C
or calculated (C)
Period Number of days
1 7 981.6 67.7 24.6 921.9 26.9 1,041.1
2 7 754.7 72.1 20.5 688.6 56.0 837.2
Total 14 1,736.3 139.8 45.1 1,610.5 82.9 1,878.3
 Examples and case studies 121

Temperature Profiles – Detached Ambientt Indoor

26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25
ary 3

ary 3

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary

uary
Janu

Janu

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr

Febr
2001

2001

2001

2001

2001

2001

2001

2001

2001
2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001

2001
Figure 5.36 Temperature profiles

Balcony
Balcony

Kitchen
Kitchen Bedroom
Bedroom

Living-Room
Living-Room Corridor
Corridor

Bedroom
Bedroom
Bathroom
Bathroom

Up
Up
Entrance
Entrance
Up
Up

Semi-detached
Semi-detached Apartment
Apartment -- 3rd
3rd Floor
Floor

Figure 5.37 Plan of the apartment; the downward direction is south-east

Table 5.93 Supplied normalized heating and global energy consumption (in kW/year)
Year Qheat, sup, normal qheat, sup, normal Qtot, sup, normal qtot, sup, normal
Normal external 16,146 206.9 21,753 278.9
15,028 < Q < 16,505 192.7 < q < 211.6 20,635 < Q < 22,112 264.6 < q < 283.5

Table 5.94 Delivered normalized heating energy consumption (in kW/year)

Year Qheat, del, normal qheat, del, normal
Normal external 13,724 175.9
Normal internal 13,655 175.1

Table 5.95 Data obtained during the energy monitoring (in kWh)
Energy consumption Hot water Cooking Energy consumption
– oil consumption consumption Qheat – electricity Qtot
Measured (M), estimated (E) M E E C M C
or calculated (C)
Period Number of days
1 7 726.3 114 32 726.3 27.5 899.8
2 7 628.1 133 52 628.1 44.5 857.6
Total 14 1,354.4 247 84 1,354.4 72.0 1,757.4
122 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

Temperature Profiles – Apartment Ambient Indoor

27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2

h2
15
6
Febr y 17
17
8
Febr y 19
20
1
21
22
3
24
4
5
26
7
8
Marc 28
Marc 1
2001 nuary 26
2001 nuary 27
2001 nuary 28
2001 nuary 28
2001 nuary 29
2001 nuary 30
2001 nuary 31
2001 bruary 1
2001 bruary 1
2001 bruary 2
2001 bruary 3
2001 bruary 4
2001 bruary 5
2001 bruary 6
2001 bruary 7
2001 bruary 7
2001 bruary 8
9
Febr y 10
11
1
Febr y 12
13
4
Febr y 14
2001 bruary 1

2001 bruary 1

2001 bruary 1

2001 bruary 1

2001 bruary 2

2001 bruary 2

ary 2
2001 bruary 2

ary 2
2001 bruary 2

h
2001 ebruary

uary

uary

uary

uary

uary

uary
uary

uary

uary

y
uar

ar

ar

ar

ar

a r
u

u
Febr

Febr

Febr

Febr

Febr

Febr
Febr

Febr
Febr

Febr
Febr

Febr
Fe
Fe
Fe
Fe
Fe
Fe
Fe
Fe
Fe
Fe

2001
2001
Ja
Ja
Ja
Ja
Ja
Ja
Ja

Fe

Fe

Fe

Fe

Fe

Fe

Fe

Fe
2001

2001
2001

2001
2001

2001
2001

2001
2001

2001
2001

2001
2001

2001
2001

2001
2001
Figure 5.38 Temperature profiles

Table 5.96 Normalization ratios for heating – NR @ 22 = 1.583, NR @ 20 = 1.575 and NR @ 18 = 1.359
Period DD @ 22 DD @ 20 DD @ 18 (n/N) CSI @ 20
Reference year 277 297 390 0.507 0.505
February 2001 175 232 287 0.583 0.293

Table 5.97 Supplied normalized heating and global energy consumption (in kW/year)
Year Qheat, sup, normal qheat, sup, normal Qtot, sup, normal qtot, sup, normal
Normal external 15,111 166.1 23,574 259.1
13,039 < Q < 15,284 143.3 < q < 167.9 21,502 < Q < 24,737 236.3 < q < 260.9

as a normalized energy consumption based on the indoor

Table 5.98 Delivered normalized heating energy consumption (in kW/year) climate.
Year Qheat, del, normal qheat, del, normal The different normalization techniques available have
Normal external 12,844 141.1 been applied in these tests (degree-day, Climate Severity
Normal internal 13,229 145.3 Index, modified utilization factor).

AUTHORS
Table 5.99 Comparison of the results of MEP and BEP methods The following authors have participated in the preparation
QHeating qHeating QGlobal qGlobal of this chapter. The monitoring and calculations have been
Building Method (kWh) (kWh/m2) (kWh) (kWh/m2) realized at a national level.
Detached MEP 16,146 206.9 21,753 278.9
house BEP 14,096 180.7 19,549 250.6 Belgium
Apartment MEP 15,111 166.1 23,574 259.1
Xavier Loncour, Belgian Building Research Institute (BBRI
BEP 17,642 193.9 26,163 287.5
– CSTC – WTCB), Boulevard Poincaré 79, B-1060-Brussels,
Belgium; e-mail: [email protected]

Table 5.100 Heat-loss factors Greece

QDelivered HLF
George Zannis and Mat Santamouris, University of Athens,
Building (kWh) Ts (◦ C) Ta (◦ C) (W/K)
Physics Department, Building & Environmental Studies
Detached house 1,284 19.9 11.3 444
Group, Panepistimioupolis, Athens 157 84, Greece; e-mail:
Apartment 1,151 20.6 12.3 413
[email protected], [email protected]
 Examples and case studies 123

Spain REFERENCES
Servando Alvarez, AICIA Universidad de Sevilla, 1. NIT 212, 1999, Code de bonne pratique pour l’installation
Departamento de Ingeniería Energética y Mecánica de des chauffes-eaux solaires. Belgian Building Research Institute,
Fluidos, Avenida Reina Mercedes, s/n-Apdo n◦ 1182, Brussels.
E-41080 Seville, Spain; e-mail: [email protected] 2. BBRI, 1999, Internal document – EPIGOON
project – werkdocument 2.007. Ontwerpnota betreffende EPN,
Sweden algemene benadering. Belgian Building Research Institute,
Brussels.
Jan Akander, Kungliga Tekniska Högskolan (KTH), 3. Persson T, 2000, Measurements of solar radiation in Sweden
Department of Building Sciences, Division of Building 1983–1998. Reports Meteorology and Climatology. Swedish
Technology, Brinellv 34, SE-100 44 Stockholm, Sweden; Meteorological and Hydrological Institute.
e-mail: [email protected]
124

APPENDIX 1
Audit form

ADMINISTRATIVE INFORMATION
Municipality House number ID number

House address:

Estate property ID Coordinates X, Y Number of buildings

Owner/administrator:

Owner address (if other than house address):

CONTACT PERSON
Building Name:
Address: Tel:
Available to be contacted (days and time):
Agreed time for audit (date and time):

Apartment Name:
Address: Tel:
Available to be contacted (days and time):
Agreed time for audit (date and time):

Audit performed (yyyy-mm-dd): ❏❏❏❏-❏❏-❏❏

Auditor name:
Audit company:

GENERAL BUILDING INFORMATION

A1 Number of buildings on the property that are to be audited? ❏❏❏

A2 Building year? ❏❏❏❏
A3 Has there been any major retrofit/conversion/extension?
1. ❏ Yes 2. ❏ No
Year of major retrofit/conversion/extension? ❏❏❏❏ ❏ Has not been performed
A4 Building location:
1. ❏ Urban area, city
2. ❏ Suburb, smaller urban areas with mainly multi-family buildings
3. ❏ Detached building area, terrace house area
4. ❏ Rural area

A5 What type of building is being audited?

Single-family units
1. ❏ Detached house
2. ❏ Terrace house
3. ❏ Semi-detached house
4. ❏ Semi-detached two-family house
5. ❏ Detached two-family house
6. ❏ Other:
 Appendix 125

Multi-family units
7. ❏ Slab block
8. ❏ Tower block
9. ❏ Balcony access block
10. ❏ Other:

A6 How many storeys are there in the building? Number of storeys ❏❏

How many storeys in the house are heated to at least 16◦ C during Number of storeys ❏❏
the winter season?
How many storeys in the house are cooled to at least 21◦ C during Number of storeys ❏❏
the summer season?

A7 How is the building utilized?

1. ❏ Residence only
2. ❏ Residence and for commercial/industrial activities

A8 Give for the building (plan drawings or ocular survey)

a) Circumference (external dimension) ❏❏❏❏ m
b) Longest length (external dimension) ❏❏❏❏ m
c) Envelope area (external dimension) ❏❏❏❏ m2
d) Total floor area (internal dimension) ❏❏❏❏ m2
whereof
e) Heated floor area (internal dimension) ❏❏❏❏ m2
whereof
f) Residential heated floor area (internal dimension) ❏❏❏❏ m2
g) Total volume (external dimensions) ❏❏❏❏ m3
A9 How many apartments are there in the building? ❏❏❏❏ apartments
What is the distribution of apartments?
❏❏❏❏ apartments ≤1 room and kitchen (r a k)
❏❏❏❏ apartments 2rak
❏❏❏❏ apartments 3rak
❏❏❏❏ apartments 4rak
❏❏❏❏ apartments ≥5 r a k

A10 Which is the main ventilation system type? (informative)

1. ❏ Natural ventilation system
2. ❏ Controlled natural ventilation system
3. ❏ Exhaust ventilation system
4. ❏ Balanced ventilation system
5. ❏ Balanced ventilation system with heat recovery

A11 Are there any recordings from the most recent tuning of the main ventilation system?
1. ❏ Yes 2. ❏ No

If single-family house, go to A14

A12 Which storey is the apartment in?

1. ❏ Bottom
2. ❏ Intermediate
3. ❏ Top

A13 Is it a gable apartment?

1. ❏ Yes 2. ❏ No

A14 Number of rooms excluding the kitchen? ❏❏

whereof bathrooms ❏
Windows and doors
126

Fill in the tables for windows and glazed parts of doors in respective sector. Sum these up and enter the total value.

A15 Sum table sector North (NW-NE) ❏ No windows

Type/Variants g-factor Curtain Lateral sector Central sector Central sector Lateral sector Glazed
factor horizontal angle horizontal angle horizontal angle horizontal angle area
-glazed m2
-glazed m2
-glazed m2
-glazed m2

Sum table sector East (NE-SE) ❏ No windows

Type/Variants g-factor Curtain Lateral sector Central sector Central sector Lateral sector Glazed
factor horizontal angle horizontal angle horizontal angle horizontal angle area
-glazed m2
-glazed m2
-glazed m2
-glazed m2

Sum table sector South (SW-SE) ❏ No windows

Type/Variants g-factor Curtain Lateral sector Central sector Central sector Lateral sector Glazed
factor horizontal angle horizontal angle horizontal angle horizontal angle area
-glazed m2
-glazed m2
Energy Performance of Residental Buildings. A Practical Guide for Energy Rating and Efficiency

-glazed m2
-glazed m2

Sum table sector West (NW-SW) ❏ No windows

Type/Variants g-factor Curtain Lateral sector Central sector Central sector Lateral sector Glazed
factor horizontal angle horizontal angle horizontal angle horizontal angle area
-glazed m2
-glazed m2
-glazed m2
-glazed m2
 Appendix 127

A16 Sum the total area of non-glazed part of external doors and balcony doors. Measure from frame-to-frame.
Total door area m2

RESIDENTS AND BEHAVIOUR (Information from the resident representative)

B1 How many person live in the apartment?

Number of adults (18 years and above) ❏❏
Number of children (0–2 years) ❏❏
Number of children (3–12 years) ❏❏
Number of teenagers (13–17 years) ❏❏
B2 How many weeks per summer season is no one living in the apartment? ❏❏ weeks
How many weeks per winter season is no one living in the apartment? ❏❏ weeks
B3 How many persons use the apartment more than 6 hours between hrs 08 and 18?
Persons during working days ❏❏
Persons during weekends ❏❏
B4 Is there a main heating system in the building? 1. ❏ Yes 2. ❏ No
At what temperature does the resident usually keep during the winter? (Occupant estimation)
Temperature Heating off
During the day when no one is at home. ❏❏ ◦ C ❏ yes
During the day when someone is at home. ❏❏ ◦ C ❏ yes
During sleeping hours. ❏❏ ◦ C ❏ yes
B5 Is there an automatic setback of the indoor temperature during the day/week at winter?
1. ❏ Yes 2. ❏ No
Set-point temperature starting hours–ending hours
❏❏.❏ ❏❏.❏❏–❏❏.❏❏
❏❏.❏ ❏❏.❏❏–❏❏.❏❏
❏❏.❏ ❏❏.❏❏–❏❏.❏❏
B6 Is there a cooling system in the building? 1. ❏ Yes 2. ❏ No
At what temperature does the resident usually keep during the summer? (Occupant estimation)
Temperature Cooling off
During the day when no one is at home. ❏❏ ◦ C ❏ yes
During the day when someone is at home. ❏❏ ◦ C ❏ yes
During sleeping hours. ❏❏ ◦ C ❏ yes
B7 Is there an automatic setback of the indoor temperature during the day/week in summer?
1. ❏ Yes 2. ❏ No
Set-point temperature starting hours–ending hours
❏❏.❏ ❏❏.❏❏–❏❏.❏❏
❏❏.❏ ❏❏.❏❏–❏❏.❏❏
❏❏.❏ ❏❏.❏❏–❏❏.❏❏
B8 When was the heating system turned on and off?
Turned on (mm-dd) ❏❏-❏❏ Turned off (mm-dd) ❏❏-❏❏ ❏ Not available
When was the cooling system turned on and off?
Turned on (mm-dd) ❏❏-❏❏ Turned off (mm-dd) ❏❏-❏❏ ❏ Not available

Heating

C1 How is the audited building mainly heated?

1. ❏ Individual boiler 2. ❏ District heating 3. ❏ Sub district heating/central boiler plant
4. ❏ Heat pump 5. ❏ Direct electrical heating 6. ❏ Other:
128 Energy Performance of Residental Buildings. A Practical Guide for Energy Rating and Efficiency

C2 Main heating type and rated power?

Rated power Age Meter no./subscription no.
Individual boiler ❏❏❏.❏❏ kW ❏❏ years ❏❏❏❏❏❏❏❏❏❏
District heating ❏❏❏.❏❏ kW ❏❏ years ❏❏❏❏❏❏❏❏❏❏
Local district heating/grpc ❏❏❏.❏❏ kW ❏❏ years ❏❏❏❏❏❏❏❏❏❏
Heat pump ❏❏❏.❏❏ kW ❏❏ years ❏❏❏❏❏❏❏❏❏❏
Direct electrical heating ❏❏❏.❏❏ kW ❏❏ years ❏❏❏❏❏❏❏❏❏❏
Solar heating ❏❏❏.❏❏ m2 collector area ❏❏ years ❏❏❏❏❏❏❏❏❏❏
C3 Are there service, tuning and operation records? 1. ❏ Yes 2. ❏ No
If yes
Individual boiler efficiency ❏❏❏% year ❏❏❏❏
District heating efficiency ❏❏❏% year ❏❏❏❏
Local district heating/grpc efficiency ❏❏❏% year ❏❏❏❏
Heat pump COP ❏.❏❏ year ❏❏❏❏
Direct electrical heating efficiency ❏❏❏% year ❏❏❏❏
Solar heating efficiency ❏❏❏% year ❏❏❏❏
C4 For individual boilers only. Estimated annual energy distributions of various energy fuels
(by responsible representative).
Do the boilers use more than one fuel? 1. ❏ Yes 2. ❏ No
Meter no./subscription no.
Light heating oil ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
Heavy heating oil ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
Natural gas ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
Electricity ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
Wood, sawdust, chips ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
Pellets ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
Coal, coke ❏❏❏% ❏❏❏❏❏❏❏❏❏❏
C5 Is there an automatic change-over between the various energy fuels?
1. ❏ Yes, temperature controls 2. ❏ Yes, timer controls 3. ❏ No
C6 Is there in the audited building other forms of supplementary heat generating sources other than heat pumps and
solar collectors?
More than one alternative is possible.
Device Frequency Quantity/ Utilisation Quantity/ Included in bills
of use/week use time season
1. ❏ Fire place (wood) m3 ❏ Yes
2. ❏ Electrical stove W hrs ❏ Yes
3. ❏ Wood heating stove m3 ❏ Yes
4. ❏ Natural gas heating stove l/h hrs ❏ Yes
5. ❏ Bottled gas heating stove l/h hrs ❏ Yes
6. ❏ Kerosene heating stove l/h hrs ❏ Yes
7. ❏ The cooker stove l/h,W hrs ❏ Yes
8. ❏ Portable heaters W hrs ❏ Yes
9. ❏ Other: hrs ❏ Yes
10. ❏ None

Heat pumps

C7 Does one or more heat pump exist? 1. ❏ Yes 2. ❏ No

C8 Are these used for
Space heating 1. ❏
Hot-water heating 2. ❏
Space and hot-water heating 3. ❏
Space cooling 4. ❏
Space cooling and hot-water heating 5. ❏
 Appendix 129

C9 Are values on supplied electricity (or natural gas) to the heat pump available?
1. ❏ Yes 2. ❏ No Meter number ❏❏❏❏❏❏❏❏❏❏❏❏

Distribution of heat

C10 Which is the type of main heat distribution in the building?

1. ❏ Water borne 2. ❏ Air borne 3. ❏ Electrical 4. ❏ None
C11 Which year was the last occasion that the heating system was tuned?
Year: ❏❏❏❏ ❏ Unknown
C12 If available from recent records, or possibility of present readings
Shunt group 1 Shunt group 2
Temperature of supply heat carrier? ❏❏❏ ◦ C ❏❏❏ ◦ C
Temperature of return heat carrier? ❏❏❏ ◦ C ❏❏❏ ◦ C
Temperature of supply heat carrier ❏❏❏ C ◦
❏❏❏ ◦ C ❏ Unknown
◦
at an outdoor temperature of 0 C?
Pressure drop in circuit? ❏❏❏ kPa ❏❏❏ kPa ❏ Unknown
Flow in circuit? ❏❏❏ l/min ❏❏❏ l/min ❏ Unknown
C13 What is the main type of heaters in the building?
1. ❏ Radiators 2. ❏ Convectors 3. ❏ Floor heating 4. ❏ Ceiling heating
5. ❏ Other:

DISTRICT HEATING

C14 Does the building have its own meter for district heating, with no other buildings included?
1. ❏ Yes 2. ❏ No

C15 Are there service, tuning and operation records? 1. ❏ Yes 2. ❏ No

If yes
District heating efficiency ❏❏❏% year ❏❏❏❏

COOLING

D1 How is the audited building mainly cooled? (more than one alternative possible)
1. ❏ Local units 2. ❏ Central unit 3. ❏ District
4. ❏ Fans (without cooling) 5. ❏ Other: 6. ❏ No cooling

D2 What is the main type of cooling equipment? (more than one alternative possible)
Type Rated power Age “Fuel”
Local units ❏❏❏.❏❏ kW ❏❏ years ❏ Electricity ❏ Natural gas
Central cooling ❏❏❏.❏❏ kW ❏❏ years ❏ Electricity ❏ Natural gas
District cooling ❏❏❏.❏❏ kW ❏❏ years
Fans ❏❏❏.❏❏ kW ❏❏ years
D3 Are there service, tuning, operation records or rated performance? 1. ❏ Yes 2. ❏ No
If yes
Local units mean efficiency ❏.❏ year ❏❏❏❏
Central cooling efficiency ❏.❏ year ❏❏❏❏
District cooling efficiency ❏❏❏% year ❏❏❏❏

Domestic hot water

E1 Is there a central domestic hot-water boiler?

1. ❏ Yes 2. ❏ No, separate boilers in every apartment/compartment
3. ❏ Hot water not available
130 Energy Performance of Residental Buildings. A Practical Guide for Energy Rating and Efficiency

E2 Are meters for the hot-water consumption for the building available? 1. ❏ Yes 2. ❏ No
E3 How is the heating of domestic water mainly done during the summer and the winter season?

Summer Winter
Water heating with conventional boiler/heat exchanger/ 1. ❏ 1. ❏
heat pump that also provides space heating
Heat pump only for water heating purpose 2. ❏ 2. ❏
Separate electrically heated water boiler 3. ❏ 3. ❏
Solar heating 4. ❏ 4. ❏
Bottled gas 5. ❏ 5. ❏
Hot water shut off 6. ❏ 6. ❏
Other: 7. ❏ 7. ❏

E4 Rated heat output of separate domestic hot-water boiler? ❏❏❏.❏❏ kW 0 ❏ Not available
Age of separate heated water boiler? ❏❏ years 0 ❏ Not available

E5 Rated heat output of central domestic hot-water boiler? ❏❏❏.❏❏ kW 0 ❏ Not available
E6 What is the set-point temperature for the hot water? ❏❏❏ ◦ C 0 ❏ Not available
E7 Size of the hot-water tank/accumulator tank? ❏❏❏❏ m3 0 ❏ Not available

Appliances

F1 What type of fuel does the stove use?

1. ❏ Electricity 2. ❏ Natural gas 3. ❏ Bottled gas 4. ❏ Other
What type of fuel does the oven use?
1. ❏ Electricity 2. ❏ Natural gas 3. ❏ Bottled gas 4. ❏ Other
F2 How many hot meals are on an average cooked each day at home?
1. ❏ Three or more 2. ❏ Two 3. ❏ One 4. ❏ A few times per week

Electricity meter information

G1 Does the building have its own electricity meter, with no other buildings included?
1. ❏ Yes 2. ❏ No

G2 Which electrical meters exist for the building?

Meter 1 Meter 2 Meter 3
❏❏❏❏❏❏❏❏❏❏ ❏❏❏❏❏❏❏❏❏❏ ❏❏❏❏❏❏❏❏❏❏
G3 The meter reads the following (more than one alternative possible)
Meter 1 Meter 2 Meter 3
Household electricity 1. ❏ 1. ❏ 1. ❏
Electrical space heating 2. ❏ 2. ❏ 2. ❏
Electrical space cooling 3. ❏ 3. ❏ 3. ❏
Hot water boiler 4. ❏ 4. ❏ 4. ❏
External lighting 5. ❏ 5. ❏ 5. ❏
Other electricity use on the property 6. ❏ 6. ❏ 6. ❏

Other electricity use on the property

G4 Are there other electrical devices or buildings on the property which use the same electricity meter?
1. ❏ Yes 2. ❏ No?
 Appendix 131

G5 Describe the object here and write, if possible, rated power and run time as a percentage for the seasons.
Rated power Run time Run time Meter/subscription no.
winter summer
Exterior lighting ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
Gutter/downpipes/water ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
electrical cables
Car heater ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
Pool heaters and pumps ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
Exterior sauna ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
Infra-red heaters ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
Outdoor grill ❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏
❏❏❏ kW ❏❏❏ % ❏❏❏ % ❏❏❏❏❏❏❏❏❏❏

SUPPLIERS

Energy supplier name, Subscript/account Meter number Fuel Unit Estimate Conv.
Address Exact factor
1
2
3
4

Water supplier name, Subscript/account Meter number Fuel Unit Estimate

Address Exact
water

No. Date Quantity Date Quantity Date Quantity Date Quantity Date Quantity
132

APPENDIX 2
Energy transmittance by glazing and
shading factors

The information in Appendix 2 is to a large extent derived Table A2.1 Typical total solar energy transmittances for the two most
from EN 832. The quantity of data has been reduced for common types of glazing. These are values for normal incidence, assuming
practical purposes: the inaccuracy in the results of the a clean surface. For other types of glazing, use certified values or national
fieldwork carried out by auditors must be minimized. default values
Glazing type g⊥
Single glazing 0.85
TOTAL SOLAR ENERGY TRANSMITTANCE Double glazing, clear glass 0.75
FOR GLAZING
Energy transmission through transparent surfaces depends
on the type of glass and the coatings. The total solar energy 2. The corresponding partial shading factor si is obtained
transmittance defined in EN 410 is calculated for solar from Table A2.2.
radiation perpendicular to the glazing, g⊥ . Some indicative 3. The shading factor s of the element under consideration
values are provided in Table A2.1. is given by:
For monthly calculations, a value averaged over all angles
⎡ ⎤
of incidence is required. The factor Fw , depending on 
4
type of glass, latitude, climate and orientation, is given s = min ⎣1, si ⎦ (A2.2)
approximately by i =1

Fw ≈ 0.7 − 0.9 (A2.1)

The sum is over the four sections.

SHADING FACTORS Shading from overhangs

Shading from an external horizon Please refer to section G 2.2 of EN832.
◦
The total viewing angle from a glazed surface is 180 ;
projected onto a horizontal surface, see the left-hand side
Curtain factors
of Figure A2.1. This angle is divided into four equal parts:
two central sections and two lateral sections: The curtain factor is the ratio of the average solar energy
entering the building with curtains to the energy that could
1. For each section, the average angle α subtended by the enter the building without the curtains. Some values are
obstacles from the horizontal plane is determined, as given in Table A2.3 for curtains placed inside and outside
shown on the right-hand side of Figure A2.1. the window.

Table A2.2 Partial shading factors s i for external obstacles

South (SE to SW) or East (SE to NE) or
Orientation of the façade North (NE to NW) azimuth West (SW to NW) azimuth
Average angle α for each For both lateral For both central For the lateral section For the central section For the two
section in degrees sections sections towards south towards south other sections
0–9 0.00 0.05 0.05 0.10 0.05
10–14 0.05 0.10 0.10 0.15 0.05
15–19 0.05 0.15 0.15 0.20 0.10
20–24* 0.10 0.20 0.20 0.25 0.10
25–29 0.1 0.25 0.25 0.30 0.10
30–34 0.10 0.30 0.25 0.35 0.15
35–44 0.15 0.35 0.30 0.40 0.20
45–90 0.15 0.40 0.30 0.45 0.25
∗ These have been slightly modified from the original values of EN832. Shading factors are valid only between latitudes of 40◦ and 50◦ .
 Appendix 133

Building

Solar collecting areas

Lateral
sectors

Central sector
towards south

Lateral sector
towards south
Central sectors
S
Figure A2.1 Left-hand side: a building as seen from above, where the view, as seen from behind each solar collecting area (window), is projected onto
the horizontal plane. Each viewing plane is divided into four sectors. Right-hand side: illustration of the angle α

Table A2.3 Curtain factors for some types of curtains, installed inside or outside the window
Optical properties Curtain factor with
Curtain type Absorption Transmission Curtain inside Curtain outside
White venetian 0.1 0.05 0.25 0.10
blinds 0.1 0.30 0.15
0.3 0.45 0.35
White curtains 0.1 0.5 0.65 0.55
0.7 0.80 0.75
0.9 0.95 0.95
Coloured 0.3 0.1 0.42 0.17
textiles 0.3 0.57 0.37
0.5 0.77 0.57
Aluminium- 0.2 0.05 0.20 0.08
coated textiles
134

APPENDIX 3
Estimated average fuel combustion
efficiency of common heating
appliances

Values of heating equipment efficiency given in Table A3.1.

Table A3.1 Values of heating equipment efficiency

Type Heating equipment efficiency (%)
Coal (bituminous) central heating, 45.0
hand-fired
Central heating, stoker-fired 60.0
Water heating, pot stove 14.5
(50 gal./227.3 litre)
Oil high-efficiency central heating 89.0
Typical central heating 78.0
Water heater (50 gal./227.3 litre) 59.5
Gas high-efficiency central heating 92.0
Typical central heating 82.0
Room heater, unvented 91.0
Room heater, vented 78.0
Water heater (50 gal./227.3 litre) 62.0
Electricity central heating, 97.0
resistance
Central heating, heat pump 200+
Ground-source heat pump 300+
Water heaters (50 gal./227.3 litre) 97.0
Wood and pellet Franklin stoves 30.0–40.0
Stoves with circulating fans 40.0–70.0
Catalytic stoves 65.0–75.0
Pellet stoves 85.0–95.0
 135

Index
Page numbers in italic refer to Figures and Tables.
accuracy 14, 19–23, 20, 21, 47, 52, 80 building size, normalization with respect to 58, 69–70
of BEP 32–3 buildings iii
and climates 21–2, 23 drying out 17–18, 19
of MEP 120 energy balance 2–3
of meters 46, 48 energy performance parameters 4
of monitoring 16, 17, 23 general information 54
of neural networks 13–14 rating ii, iii, 71, 72, 75
air change rates 8, 9, 10, 10, 16, 22, 51 size 57, 58, 69–70
air conditioning 96, 97, 99 technical data 1–2, 8, 33
air leakage 41, 114 see also case studies; occupied dwellings; unoccupied buildings
annual total energy use 51, 52
appliances see external appliances; household appliances calculation methods 1, 14–15
area-specific global energy 69 carbon dioxide (CO2) concentration 14
ARMA (AutoRegressive Moving Average) model 15 case studies
audits i, ii, 32, 33, 52–6, 74 Belgium (BEP) 78, 79–84, 80, 81, 82, 83, 84, 85, 86
for BEP 35–6 Belgium (MEP) 106–9, 107, 108, 109
forms 52–3, 53–4, 124–31 BEP 78–104
for MEP 42, 44, 46, 58 Greece (BEP) 78, 97–104, 101, 102, 103, 104, 105, 106
AutoRegressive Moving Average model see ARMA Greece (MEP) 116–20, 120, 121, 122
MEP 106–20
base indoor temperatures 59, 61, 67 selection of buildings for 78–9, 78
behaviour of occupants i, 10, 14, 51, 58, 96, 97 Spain (BEP) 78, 94–7, 95, 96, 97, 98, 99, 100
auditing 33, 54–5, 127 Spain (MEP) 114–16, 119, 120
evaluating 12 Sweden (BEP) 78, 85–94, 87, 88, 89, 90, 91, 92, 93, 94, 95
influence of 23, 62 Sweden (MEP) 109–14, 110, 111, 112, 113, 114, 115, 116, 117, 118
normalizing 57 CEV (Spanish Energy Labelling for Social Housing) project 65–6,
standard occupants 58, 67 65, 66
Belgium iii, 9, 20, 20, 21, 66, 66 climate 4, 21–2, 31
BEP case studies 78, 79–84, 80, 81, 82, 83, 84, 85, 86 Belgium 80, 80, 81
climate 80, 80, 81 Greece 98–9, 101, 102
MEP case studies 106–9, 107, 108, 109 microclimate 120
BEP (Billed Energy Protocol) iii, 31–4, 32, 33, 35–42, 54, 71, 73, 79, Spain 95, 96
104–5 Sweden 85–6, 87
case studies 79–104 see also CSI
compared with MEP 44, 120–1 climate chambers 16, 16
conditions for 34–5 Climate Severity Index see CSI
evaluation 104–5 climatic data 35, 46, 51, 58, 72, 73, 76
limitations 34, 34 measuring 44, 46
MEP as alternative to 34, 34, 42 sources 8, 10, 34, 51, 58
normalization of output 58 see also outdoor temperatures
protocol forms 42, 42, 43, 50 co-operative living societies, Sweden 89–91
BEVA (Building Energy Vector Analysis) 15 CO2 (carbon dioxide) concentration 14
Billed Energy Protocol see BEP coal/coke 49
billed (supplied) energy 28, 28, 35, 36, 42, 43, 57–8, 74, 76 common spaces 91
measurement 42, 47–8 common walls 18–19, 18
seasonal 40–1 comparison scenarios 71, 76
billing systems iii, 78 confidence intervals 19, 20, 23, 52
Belgium 79, 81, 82–3 control, of heat power 14
Greece 97, 101, 103 conversion factors, fuels 2, 2, 3, 35, 38, 45
Spain 94–5, 96, 97 cooling see space cooling
Sweden 85, 86, 89, 92 cooling CSI (Climate Severity Index) 65, 65, 66
bills see billing systems; energy bills cooling season 71, 72, 96, 97
biomass fuels 23, 49, 85, 86, 87, 88 cooling systems i, 96, 97, 99
see also wood costs, of energy rating 31, 32, 42
BLC (building loss coefficient) 11, 12 CSI (Climate Severity Index) iii, 59, 64–7, 65
bottled gas 49 Greece 98–9, 101, 101, 103–4, 118
buffer spaces 12 curtain factors 62, 132, 133
Building Energy Vector Analysis (BEVA) 15
building load simulations 14, 15 DDM (degree-day method) iii, 59–60, 67–8, 95, 101, 103–4
building loss coefficient (BLC) 11, 12 used in case studies 84, 89, 94, 99
building mass discharge 12 degree-days iii, 3, 66, 72, 73, 76, 85, 96–7
building score 71, 72, 75 degree-days 15/15 concept 79–80
136 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

degree-hours 2 from retrofitting 2, 3, 4, 4

degree-months 62 Swedish measures i, 1
delivered daily base energy use 41–2 Energy Signature see ES
delivered energy 28–9, 28, 29–30, 34, 50, 57–8, 74, 76 energy use i–ii, 15, 74
allocation to specific energies 34, 36, 37, 43 annual total 51, 52
seasonal 40–1 see also energy consumption; hot water; household appliances;
total 71 lighting; space cooling; space heating
delivered space cooling 41 environmental issues ii
delivered space heating 40–1 equivalent area-specific space-cooling energy 69–70
Denmark, energy rating ii errors 10, 14, 62
diesel oil 94, 95, 97, 99, 101 ES (Energy Signature) method 42, 50, 50–2, 52, 74, 94, 95
difference, indoor/outdoor temperatures 5, 5, 9, 11, 14, 21, 51, 52, accuracy 14, 46
59 estimates 7, 49
differential equations 15 external energy use 39–40
district heating 49, 55, 85, 91, 129 fuel consumption 38–9
domestic load, predicting 14 normalized energy consumption 52
drainage losses 3, 5, 12 see also BEP; MEP
drying out of buildings 17–18, 19 EUROCLASS project iii, 64, 66, 71
dynamic energy balance 6–7, 11 see also EUROTARGET software
EUROTARGET software iii, 71–2
database (constants) 71, 72, 73, 76
E-box 7–8 input data 71–2, 72, 73, 74, 77
EB (Energy Barometer) i–ii, 6–8, 8, 15 results (output) 72, 72, 73, 75, 77
EEM (Energy Efficiency Mortgages) ii Excel (Microsoft Excel) see EUROTARGET software
effective building mass 11 experimental protocols ii, iii, 26–9, 31–2, 32
effective solar gain area 11 see also BEP; MEP
efficiency external appliances 37, 92, 130–1
heating/cooling systems 2, 2, 3, 51, 134 energy for 28, 29, 30, 31
see also energy conversion systems external climate 45, 50
electricity normalization with respect to 57, 58, 59–67, 68
meters 56, 94, 97, 101 see also outdoor temperatures
monitoring 23, 48 external energy use 39–40
ELIB project 52–3 external spaces 31, 91, 92
emissions to environment i
end-users see occupants
fuel consumption, estimates 38–9
energy
fuel conversion factors 2, 2, 3, 35, 38, 45
consumption see energy consumption
fuel-oil 94, 95, 97, 99, 101
delivered see delivered energy
fuels 29, 36
global 57–8
conversion factors 2, 2, 3, 35, 38, 45
primary 28, 28
for cooking 56, 79
specific see specific energies
multi-use 79, 84
supplied see supplied energy
see also energy bills; energy conversion systems; hot water;
household appliances; lighting; space cooling; space heating gA values (solar gains) 1, 9, 19–21, 20, 21, 23
energy balance 2–3, 3, 6–7, 11 see also solar gains
Energy Barometer (EB) i–ii, 6–8, 8, 15 gain-loss ratio 60, 61
energy bills i, i–ii, iii, 1–2, 34, 36, 40, 78 gas see bottled gas; natural gas
in BEP 32, 34, 35, 36, 36, 37 gauges, location 46–7, 47, 53, 116, 118
in MEP 32, 42, 45–6, 50, 51 glazed area 51, 54, 80, 96, 97, 126–7
see also billing systems glazing 15–16, 96, 97, 126–7
energy consumption i, iii, 3, 51 solar energy transmittance 51, 132, 132
measurement models 4, 4–5, 5 global energy 57–8
measurements 9, 10, 10 Greece iii, 97
normalization 50–2, 52, 57 BEP case studies 78, 97–104, 101, 102, 103, 104, 105, 106
simulations of 14 climate 98–9, 101, 102
theoretical models 4, 5–6 MEP case studies 116–20, 120, 121, 122
energy conversion systems 33, 36
efficiencies 35, 46 heat gains
in BEP 34, 35, 36, 36, 38 from appliances 3, 5
in case studies 82, 84, 88, 92, 98, 100, 103, 105, 112, 116 metabolic 3, 5, 29, 30, 31, 34, 45, 51
in MEP 42–3, 45, 46 see also solar gains
energy crisis (1970s) i, 2, 85 heat losses 12, 41, 60, 61, 67
energy demand, predicting total 14 see also transmission losses; ventilation
energy efficiency i heat pumps 4, 7, 55, 96, 97, 99, 128–9
Energy Efficiency Mortgages (EEM) ii heat-loss factor methods of normalization 68–9
energy flow, idealized model 28–31, 28 heat-loss factors 3, 5, 54, 59, 60, 92–4, 95, 104, 120
energy performance 4 estimating 42, 45, 50–2, 52, 111–12, 113, 114
energy rating i–iii monitoring period 42, 44–5, 45, 46, 67, 112
costs of 31, 32, 42 optionally estimated in BEP 34, 36, 92
see also EUROCLASS heated floor area 57, 58, 69–70, 73
energy savings heating see space heating
 Index 137

heating and cooling degree-day method 59–60, 67–8 conditions for 39, 45–6
heating CSI (Climate Severity Index) 65, 65 measurements for 32, 42–4, 58
heating season 71 normalization of output 58, 67, 68
length 3, 5, 6, 51, 74, 79–80, 96, 97 processing of data 49–52
Sweden 2, 85–6 pros and cons 44–5, 45
heating systems i, 55 metabolism, energy from 3, 5, 29, 30, 31, 34, 45, 51
efficiency 2, 2, 3, 51, 134 metered energy see billed energy
set-point temperatures 60, 68, 76, 86, 88, 92, 104 meters 4, 44, 47–8
Heating, Ventilation and Air Conditioning (HVAC) systems 14 electricity 56, 94, 97, 101, 130
heavy monitoring 13, 44 water 49
HERS (Home Energy Rating System) ii methodologies iii
hot water 37–8, 37, 49, 55–6, 109, 111, 129–30 microclimate effects 120
consumption of 35, 50, 91 Microsoft Excel see EUROTARGET software
energy consumption 2, 2, 67, 79, 91, 92, 111 modal analysis 15
energy for heating 5, 28, 29, 30–1, 31, 51 modified gain-loss ratio 61
solar heating 82, 84, 101, 104, 119 modified utilization factor see MUF
household appliances 49, 56, 91, 92 Monitored Energy Protocol see MEP
audit 130 monitoring 4, 11, 91
energy consumption 2, 2, 8, 67 of electricity 109
energy for 28, 29, 30, 31, 37, 38, 51 Energy Barometer 6–8, 8
external 130–1 of energy costs i–ii, 6
heat gains from 3, 5 equipment for 22–3, 22
see also external appliances; heating systems heavy 13, 44
humidity 47 light 8–9, 15, 19
non-occupied buildings 16–19, 16, 17, 18, 20–1, 21
occupied buildings 19–23, 19, 20, 21
HVAC (Heating, Ventilation and Air Conditioning) systems 14
of thermal performance 16–23, 17, 18, 19, 20, 21, 22
indoor climate 34, 45, 57, 67–9
see also EB; MEP; PSTAR; sensors; STEM
indoor temperatures 2, 4, 12, 13, 51–2, 54–5, 58, 67–9
monitoring periods 9, 23, 31, 32, 67
and accuracy 20, 20
for MEP 42, 44, 46, 49–50, 58
difference from outdoor 5, 5, 9, 11, 14, 21, 51, 52, 59
Monte-Carlo simulations 19–21, 20, 69
monitoring 4, 8–9, 10, 10, 11, 16, 42, 46–7, 46, 47, 67
MUF (modified utilization factor) method 59, 60–2, 67, 68–9, 79
normalization 57, 58, 67–9, 107, 109, 111
pre-study 62–4, 63, 64
inertia ratios 72, 73
used in case studies 86, 87, 88, 92, 94, 95, 111, 114, 89
infiltration air 12
multi-use of fuels 79, 84
inhabitants see behaviour of occupants; occupants
insolation hours 66
insulation i, 3, 23, 53, 80, 96, 97 national parameters 57
of envelope 114 natural gas 23, 49, 94, 95, 97
thermal insulation level 17–18, 18 neural networks 12–13, 13, 15
internal base temperature 59, 61, 67 applications 13–14
internal climate 50 non-billed energy 31, 35, 42, 47, 50, 55, 88
normalization with respect to 57, 58, 67–9 in MEP 44, 45, 88
see also indoor temperatures sources of data 33, 34, 35, 55
internal gains iii, 12, 19, 19, 23, 59, 60, 65, 68 see also wood
see also metabolism non-electric appliances 38–9
internal mass storage 6, 9, 11 non-heating season 3, 5, 51
Ireland, energy rating ii non-occupied buildings, monitoring 16–19, 17, 18, 20–1, 21
irradiation see solar irradiation normalization i, ii, iii
of BEP and MEP output 58
in case studies 79, 79
leakage area 11, 12 degree-days correction 79–80
light monitoring 8–9, 15, 19 of energy consumption 20, 20, 23, 32, 50–2, 52, 57, 79–80
lighting i, 37, 38, 56, 67, 91, 92 with respect to external climate 2, 4, 57, 58, 59–67, 68, 71, 85–6
energy for 28, 29, 30, 31 with respect to heated floor area (building size) 58, 69–70
liquified petroleum (bottled gas) 49 with respect to internal climate 57, 58, 67–9
with respect to operating conditions 71
mean base loads 50
mean indoor temperature, assessing 51–2 occupants i–ii, 6, 38–9
measurement models, for energy consumption 4, 4–5, 5 behaviour see behaviour of occupants
measurements information for i, 58, 67
continuous 9 standard 58, 67
Energy Barometer 7–8 see also co-operative living societies
for MEP 32, 42–4, 58 occupied dwellings, monitoring 19–23, 19, 20, 21
of natural gas use 23, 49 oil, measurement of use 48–9
see also MEP; monitoring oil crisis (1970s) i, 2, 85
MEP (Monitored Energy Protocol) iii, 31–2, 32, 33, 44, 54, 71, 73, orientation 132, 132, 133
86–7 outdoor temperatures 2, 8, 9, 12, 13, 72
as alternative to BEP 34, 34, 42 difference from indoor 5, 5, 9, 11, 14, 21, 51, 52, 59
case studies 106–20 monitoring 4, 5, 10, 16, 47, 47
compared with BEP 120–1 see also climatic data
138 Energy Performance of Residential Buildings. A Practical Guide for Energy Rating and Efficiency

parameter estimation 14–15 solar radiation 9, 14, 51, 61, 62, 65, 68
PASLINK test cells 16, 16, 17, 18 monitoring 10, 16, 22, 22, 47, 47, 85
people, metabolic energy from 3, 5, 29, 30, 31, 34, 45, 51 Sweden 87
PFT (PerFluor Tracergas) 8, 10, 10, 15 Somogyi, Zoltan, PhD thesis 19–21, 19, 20, 21, 69
see also tracer gas space cooling ii–iii, 15, 36, 37, 55, 69, 129
PLEIADE project 16–19, 17, 18 delivered 41
prEN ISO13790 60–2 energy for 28, 29, 31, 62
primary energy 28, 28 reference conditions 72
Primary- and Secondary-Term Analysis and Renormalization requirements 64
(PSTAR) 10–12, 15 space heating 36, 37, 55
PRISM model 15 annual demand 13–14
protocol forms, BEP 42, 42, 43 audit 127–9
protocols see BEP; experimental protocols; MEP delivered 40–1
pseudo-solar gain factor 9 energy for 8, 28, 29, 31, 57, 69
PSTAR (Primary- and Secondary-Term Analysis and normalization iii, 61–2, 68
Renormalization) 10–12, 15 reference conditions 72
requirements 64, 65
rating score (EUROCLASS) 71, 72, 75 Spain ii, iii, 65–6, 65, 66, 94–5
RC networks 19, 19, 69 BEP case studies 78, 94–7, 95, 96, 97, 98, 99, 100
RECS (Residential Energy Consumption Surveys) 53 MEP case studies 114–16, 119, 120
reference building 75 Spanish Energy Labelling for Social Housing project (CEV) 65–6,
reference climates 58, 59, 68, 69 65, 66
reference conditions 72, 73, 74, 76 specific energies 34, 36, 37, 47, 50, 57, 67
reference rating distribution 72 fuels used 29
reference years 59, 61, 65, 95, 98 measurement 42, 114
renormalization 11, 12 sub-metering 45
Residential Energy Consumption Surveys see RECS types 28, 29–31
retrofitting i, 1, 14, 54 standard occupants 58, 67
energy savings from 2, 3, 3–6, 4 static energy balance 6, 7, 9, 11
STEM (Short Term Energy Monitoring) 10–12, 11, 15
sub-metering 34, 39, 40, 43–4, 45, 54, 55
sampling frequencies 11, 46, 48, 91, 116, 118 of energy for hot water 37–8, 55
Save Belas project 43 of gas 49
Save HELP (Save House Energy Labelling Procedure) project 8–10, SUBMET model 15
8, 10, 15, 19 supplied (billed) energy 28, 28, 35, 36, 43, 58, 74, 76
seasonal efficiency 38 measurement 42, 47–8
seasons 50 seasonal 40–1
partitioning energy use 40–1, 55 total 71
see also cooling season; heating season Sweden iii, 85
sensitivity analysis (MUF) 62 BEP case studies 78, 85–94, 87, 88, 89, 90, 91, 92, 93, 94, 95
sensors 4, 8–9, 10, 10, 22 climate 85–6, 87
Energy Barometer 7–8, 7 ELIB project 52–3
location 46–7, 47, 53, 116, 118 energy saving measures i, 1
for MEP 42, 44, 46 MEP case studies 109–14, 110, 111, 112, 113, 114, 115, 116, 117,
temperature 109 118
set-point temperatures 58, 61, 67, 114 university projects 1–6
of heating systems 60, 68, 76, 86, 88, 92, 104 use of neural networks 13–14
of hot water 56, 111
setback temperatures 96, 97, 106, 109
shading control 41, 67 tap hot water see hot water
shading factors 51, 62, 132, 132, 133 temperatures
Short Term Energy Monitoring (STEM) 10–12, 11, 15 indoor/outdoor difference 5, 5, 9, 11, 14, 21, 51, 52, 59
simulations 19, 68–9 internal base temperature 59, 61, 67
building load 14, 15 monitoring equipment 22, 22
Climate Severity Index 67 patterns 50
Monte-Carlo 19–21, 20, 69 see also indoor temperatures; outdoor temperatures
total energy consumption 14 test reference year (TRY) 65
single-zone buildings 8, 12 theoretical models, energy consumption 4, 5–6
software (EUROCLASS) see EUROTARGET software thermal bridges 18, 51
solar absorption 11 thermal inertia 6, 9, 11, 118, 119
solar apertures 30, 51, 61, 62 thermal insulation level 17–18, 18
solar energy 34, 45, 103, 104 thermal mass 118, 119
solar energy transmittance, glazing 132, 132 thermal monitoring see monitoring
solar gains iii, 3, 11, 12, 19, 19, 21, 59, 60 thermal networks 15
and errors 10 thermal performance
minimizing 16, 17 non-occupied buildings 16–19, 17, 18
see also gA values occupied buildings 19–23, 19, 20, 21
solar heating 55 time constants 60, 61, 61, 62, 72, 73, 88, 92, 95
of hot water 82, 84, 101, 104, 119 TMY (typical meteorological year) 65
solar irradiation 3, 4, 29, 30–1, 31, 52, 62, 89 total delivered energy 71
on a horizontal surface 6, 51, 72, 73, 76 total energy consumption, simulations 14
 Index 139

total energy use, annual 51, 52 USA (United States of America) ii, 53
total supplied energy 71 see also PSTAR; STEM
tracer gas 8, 10, 10, 15, 16, 22 utilization factors iii, 60, 62, 68
transmission factor 51, 132 see also MUF method
transmission heat loss coefficient see UA values
transmission losses 5, 16–18, 17, 19, 19, 41, 61, 114 variable of performance 71
North Europe 21 ventilation i, 3, 54, 110
TRY (test reference year) 65 by opening windows 99, 118, 119
typical meteorological year (TMY) 65 losses by 5, 9, 16, 17–18, 17, 19, 19, 41, 61
rates 8, 9, 10, 10, 16, 22, 51
U values 2, 15–16, 23 VHL (Virtual Housing Laboratory) 7
UA and gA method 42
UA values 1, 9, 19–21, 20, 23 water consumption 50, 55, 87, 91
UK (United Kingdom), energy rating ii water heating see hot water
uncertainties 10, 14, 23, 34–5, 39, 62, 116, 118 water vapour balance 16, 17
university projects, Sweden 1–6, 2, 3, 4, 5 windows i, 51, 54
unoccupied buildings, monitoring 16–19, 16, 17, 18, 20–1, 21 see also glazing; solar apertures; U values
UP1 university project 2–3, 2, 3, 4 wood 88
UP2 university project 3–6, 4, 5 consumption 86, 87, 89, 109, 110–11, 110, 113