AMI System Using DLMS

A White Paper

Kalki Communication Technologies Limited

#147, 5th Main, HSR Layout, Sector 7
Bangalore 560102, India
E-mail: [email protected]
Phone: 91-80-4052 7900
Web: www.kalkitech.com

(This white-paper was presented at Metering India 2011, a prestigious metering event in India
conducted by IEEMA)

Abstract
Advanced Metering Infrastructure is the key terminology which is in spot light especially in Energy
Industry for last couple of years. Power sector reforms happening in most of the countries, like
APDRP in India, has been focused on distribution loss optimization and thereby reducing the AT&C
losses for improving the profit of the utilities. The need for communicable and possibly smart meters
with open communication protocol with smart communication infrastructure and scalable head end
system has attained significance. Only a scalable, flexible and open end to end Smart metering
system will offer long term benefits, not only for system designers, operators and maintainers, but
also to customers. This paper identifies some of the essential requirements for building large scale
AMI system.

Keywords
AMI/AMR, DLMS, Smart Meters

Introduction
Study shows that by 2030 total energy consumption in India will be greater than 950,000MW. India
has a unique problem of having the highest amount of aggregate technical and commercial losses in
power distribution. Generation plants coming on stream has received a boost with UMPP's, but still
we are behind target in our 5 Year plans. Without action, any energy shortage will become markedly
worse. Key to managing the situation is our ability to accurately measure usage, thereby taking action
on losses and thefts. Measurement is one of the first steps toward effective management and refining
consumer behavior. This paper highlights some of the advantages of DLMS/COSEM standard suite
(IEC 62056 / EN 13757-1) which is one of the most widely accepted international standards for utility
meter data exchange in AMI system.

Smart Meters
Smart Meters are advanced meters that collects energy-use data and transmit and receive data
to/from the utility system. Electric energy-use will be recorded on every hour or 15 minutes this
depends upon the regulatory requirement. Use of smart meters enables users and suppliers to make
use of following functionalities:
a. Time of Day usage pattern and storage at discrete intervals from 15 minutes, 1 hour, to specific
time blocks within the day. These data can be used for consumption information to central
database for various purposes like billing, ABT, and analysis purposes using wired or wireless
communication infrastructure.
b. Demand response refers to the reduction of customer energy usage at time of peak usage inorder to help address system reliability, reflect market conditions and pricing and support
infrastructure optimization or deferral.
c.

Improved accuracy of forecasting energy demand at different times of the day. Setting flexible
tariffs that measure consumption over time or otherwise called as Dynamic pricing. Different types
of dynamic pricing are: Time of usage pricing, Critical peak pricing, and Real time pricing. Critical
peak pricing and Real time pricing are of two types: day ahead pricing or hour-ahead pricing.
Same smart meters can be used for multiple suppliers whether they are following pre-paid or
credit based billing.

d. Two way interface metering for renewable integration and roof-top solar and other types of
generation.

Copyright Kalki Communication Technologies Ltd. All Rights Reserved

Communication Protocol - DLMS COSEM

There are a lot of standard metering protocols which can carry out these functionalities. The most
commonly used standard for meter data exchange is the FLAG protocol, standardized for the
purposes of electricity metering as IEC 61107 by IEC TC 13 and DLMS/ COSEM. Other protocols
widely used are Euridis protocol (used mainly in France) over twisted pairs, standardized by IEC TC
13 as IEC 62056-31 for electricity metering, MBUS standardized by CEN TC 294 as EN1434-3 (future
EN 13757-2 and -3), IEC 60870-5-102 for transmission of integrated totals, in and between
transmission and distribution stations, standardized by IEC TC 57, in North America the ANSI C12.18
(optical port), C12.19 (utility tables) and C12.21 (communication through telephone modems).
DLMS/ COSEM stands out from the list and its particularly suited to meet the needs of the liberalized
energy markets because of the following advantages:
Companion Specification for Energy Metering (COSEM) provides a layer for processing application
layer request and responds in a communication profile independent manner. This enables COSEM
application layer to support many communication profiles for example HDLC for serial, Transport layer
for IPV4 for TCP/IP networks.
Data layer defines interface classes for different kinds of data. Interface classes are broadly classified
into four types: data storage, time and event bound control, access control, and communication
channel set-up. Data storage classification includes data(IC: 1) registers(IC: 3), demand registers(IC:
5), generic profiles(IC: 7), utility tables(IC: 21), register table(IC: 61) while time and event bound
control classification defines interface classes for clock (IC: 8), single action schedule(IC: 22), special
days table (IC: 11), script table (IC: 9), schedule(IC: 10) . This model supports future expansion by
allowing manufacture specific instances, attributes, methods without changing the lower layer
services. The specific classes can be used to instantiate multiple instances in typical OOPS
terminology and each instances of same classes will be unique identification code called OBIS codes.
OBIS codes are standardized identification system that identifies all data severed by meter
application. Coding format is a six digit sequence denoted as a.b.c.d.e.f, in which each digit has its
own significance [2].
DLMS/ COSEM standard helps users to use a generic driver which enables to communicate with
various meter types from different manufactures. It is to be noted that for using other old protocols like
IEC61107, a specific driver is required for each type of meters.
This unique combination of features, not available in any other metering protocols known presently,
supports the business processes of the various stakeholders on the liberalized markets, supports
innovation and competition, and drastically reduces system life cycle costs.

Communication
Choosing the communication infrastructure is another key aspect for making a stable AMI system,
which is reliable and scalable. As far as communication options are concerned there are various
methodologies defined below:
a.
b.
c.
d.

Wired Communication (Ethernet/serial) adoption

Cellular technologies
RF
Power line

There are technological and commercial advantages and disadvantages for all the options mentioned
above. Wired system will be more reliable and scalable. However, the cost will be much higher
compared to other technologies. Selection of the wireless technology depends upon various other
factors like location, distance, geographical layout. But here again, the protocol selected for AMI
purpose will be dependent on the communication media selected by the utility.
One of the strengths of DLMS/ COSEM is that application modeling is well separated from the
communication profiles. Therefore, application data can be easily transported over various media.
The DLMS/ COSEM standard suite has been developed based on two strong and proven concepts:
3

Copyright Kalki Communication Technologies Ltd. All Rights Reserved

object modeling of application data and the Open Systems Interconnection (OSI) model. This allows
covering the widest possible range of applications and communication media. As the interface model
is completely independent from the communication media, a wide choice of media can be used,
without ever changing the model and the data management application of the data collecting system.
While today, serial interfaces are supported, using DLMS/ COSEM over the Internet is already
planned. At this time the protocol stack defined in the Green book allows to use DLMS/COSEM
through direct connection via an optical or electrical port, via switched or leased telephone lines and
over the GSM/ GPRS network.
As the COSEM application layer is separated from the lower layers, it is easy to define any lower
layer protocols, based on OSI to support any communication media.

Smart meter data acquisition

Meter data acquisition system includes two major components, Metering head end system and Meter
data management system. Metering head end system is a combination of software and hardware for
collecting data from huge number of meters/ data concentrators installed in the field using push/ pull
mechanism. Its sole responsibility is to handle the connection with the remote nodes and collect the
data and provide it to the meter data management system. It is also expected to provide statistics and
reporting information on the meter data collection, and alarms any events that require manual
intervention, like the meter being faulty; modem faulty. Meter data management is the application
software which provides long term storage and management which include the data validation,
reporting, notification. Data stored will be used by the enterprise application for energy management,
billing, outage management, and customer service system or by any application which helps utility to
stream line its business process.
An AMI system typically will have to manage from medium to large pool of meters, and could easily
go up to 1-10 Million meters. The most critical issue in such AMI systems is the scalability and
distributed network capability of the system. High scalability can be achieved by designing a system
with N-tier architecture. Since application as a whole runs sequentially, bottle neck in one of the tier
might cost a lot in performance and scalability point of view. However this limitation can be overcome
using distributed architecture.
Here we propose a distributed message queue based architecture, with distributed meter reading
modules, and a scalable messaging interface that concentrates the data at a central location. In
Message queuing architecture each and every major components of head end system are segregated
in to separate application and interconnected using a central message queue, which allows loose
coupling between the disparate systems. This also makes the system simple to maintain. Messages
are queued asynchronously between applications and systems, thereby enabling each application to
run in same hardware or in different hardware. Multiple providers can post messages to a queue and
there can be multiple message consumers attached to the single queue. Messages can be
successfully submitted to a queue even if the messages consumers for that queue are not running or
are unreachable, thereby enabling the total reliability of the system. Each application components can
act as both provider and consumer, thereby enabling two way communications for smart grid
requirement.
A simple metering head end system consists of a central message queue, meter data reading
module, and a central database for storing configuration details and data collected. Meter Data
Reading (MRM) module will handle collection of data from the meter. This can be a single application
with multiple protocols or separate application which is having different protocol drivers. MRM will be
deployed based on the requirement of number of meter data to be collected and frequency of data
collection etc. MRM will be assigned with adequate number of meters that it can handle which also
depends upon the hardware performance capability. A predefined, performance evaluated
deployment can be resorted to, or the MRM can be configured to report its running state and load on
the system, and the Administrator module can re-assign the meters to a less loaded system.
Administrator can also prioritize each queue depending on the requirement, for example demand
scan and connection/disconnection from meter data management system will be configured for high
priority queue, so that MRM can suspend all other activities and take care of the high priority
commands. User will be able to configure each data under different priority so that consumer can
4

Copyright Kalki Communication Technologies Ltd. All Rights Reserved

process the high priority queue before moving to the low priority data. This feature helps utility to
provide priority for the demand response/ remote switch off of the loads in case of emergency.
Distributed system invariably tend to be heterogeneous, that is composed of several operating
system, running application developed in several languages. Another advantage of head end system
using distributed architecture is that its very easy to incorporate meter reading software from the
meter manufactures like MIOS using customizable MRM adapter for proprietary protocols. Typical
MIOS MRM adapter has two major components, one for reading the meter data and other for parsing
the CDF file and pushing the data to database using message queue.
Distributed head end system can run in single hardware having single OS, single hardware using
multiple OS in a virtualized environment and in multiplehardware. Virtualization is software technology
which uses single server hardware and divides it in to different virtual machines, which enables user
to install multiple OS in a single machine and run different components of head-end in separate OS of
same machine. Main advantages of virtualization are server consolidation, reduced power and
cooling, ease of deployment and administrator, and high availability. Even some of the processor
available in the market from major vendors supports hardware-assisted virtualization/ accelerated
virtualization in which each OS executes in complete isolation thereby improving the performance of
the whole system. However in case of head end components to run in distributed hardware
environment, network should be designed in such a manner that the data transfer between different
components in head end system should not saturate the network and proper firewall or encryption
mechanism for data should be provided for additional security.
MDM software is intended to provide all the features and functionality utilities need for meter-to-bill
value chain, settlement and forecasting. Smart MDM will support all aspects of C&I, roll-out support,
validation, analysis, balance reports, settlement and preparation of data for billing along with allowing
utilities for configure regulatory requirements and strategies. Software will also have web-services
based on SOA for integrating to CIS and back office system. MDM system will have flexible tool for
handling calculation based on a time series, either metered or as a result of pre-calculations and
calculated data will be stored or exported to third party systems like CIS, GIS, SCADA, ABT or
accounting systems[4].

Conclusion
Utility who finds out new analysis strategies with the meter data, has found themselves in the
transforming process of adopting a smarter AMI solution for increasing the efficiency and savings over
a period of time. Smart AMI System ensures to deliver a scalable, open and common standard for
metering. This will enable utilities and end users to save time, and money by better control over the

energy use and costs.

Reference
1. Pathway for energy and climate changes.pdf
2. DLMS standard documents
3. Powel Meter Data Management System, Powel ASA

Abbreviation
ABT
AMI
AMR
APDRP
CIS
GIS
COSEM
CDF
DLMS

Availability Based Tariff

Automated Meter Infrastructure
Automated Meter Reading
Accelerated Power Development and Reform Program
Customer Information system
Geographic Information System
Companion Specification for Electricity Metering
Common Data Format
Distribution Line Message Specification

Copyright Kalki Communication Technologies Ltd. All Rights Reserved

GPRS
GSM
IEC
IC
MRM
MIOS
OOPS
OS
OSI
MDM
RF
SOA
TC
UMPP

General Packet Radio Service

Global System for Mobile communication
International Electro-technical Commission
Interface Class
Meter Reading Module
Meter Inter Interoperability Solution
- Object oriented Programming Structure
Operating System
- Open System Interconnection model
Meter Data Management
Radio Frequency
Service Oriented Architecture
- Technical committee
- Ultra Mega Power Plants

Copyright Kalki Communication Technologies Ltd. All Rights Reserved