Energy and Buildings 51 (2012) 255–262

Contents lists available at SciVerse ScienceDirect

Energy and Buildings

journal homepage: www.elsevier.com/locate/enbuild

Impacts of ice storage on electrical energy consumptions in office buildings

Fakeha Sehar ∗ , Saifur Rahman, Manisa Pipattanasomporn
Virginia Tech-Advanced Research Institute, Arlington, VA, 22203 USA

a r t i c l e i n f o a b s t r a c t

Article history: Cooling demand constitutes a large portion of total electrical demand for office buildings during peak
Received 23 October 2011 hours. Ice storage technology can help shift this peak cooling demand to off-peak periods. This research
Received in revised form 10 January 2012 analyzes the chiller energy consumption of conventional non-storage and ice storage cooling systems
Accepted 8 May 2012
for large and medium-sized office buildings in diverse climate zones. Demand Response Quick Assess-
ment Tool (DRQAT) has been used to model and simulate large and medium-sized office buildings. The
Keywords:
construction and weather files in DRQAT have been modified to incorporate construction standard and
Ice storage system
weather data for the cities representing the diverse climate zones. Results indicate that the chiller energy
Conventional non-storage system
Large and medium sized office buildings
consumption for non-storage and ice storage systems depends highly on climatic conditions. Climate
Climate zones zones with hot summers as well as small day and night temperature variations show higher chiller energy
consumptions. The marine climate zone has the lowest chiller energy consumption. The cold/humid cli-
mate zone has higher chiller energy consumption than the cold/dry and very cold climate zones. The
cold/dry and very cold climate zones have comparable chiller energy consumption. Research findings as
presented in this paper will help utilities and building owners to quantify the benefits of installing ice
storage systems in office buildings located in different climatic conditions.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction reliability. Ice storage systems can permanently shift the energy
use from peak periods to off-peak night periods thereby slashing
Total electricity consumption in office buildings is dominated the peak electricity demand. This is achieved by charging ice stor-
by cooling. Internal loads generated by building occupants, light- age by operating chillers at off-peak periods and discharging the
ing and computers impose a constant cooling load over the entire ice storage by melting ice during peak periods to meet the build-
year. During summers cooling requirement is further increased due ing cooling demand. This raises the possibility to reduce or even
to additional loads such as solar heat gain through windows, con- eliminate the chiller operation during peak periods.
duction and infiltration through building envelope and ventilation Authors in [1] provide an extensive review of various types of
requirement to maintain indoor air quality. During the daytime due thermal energy storage techniques currently available. Authors in
to high cooling demand, the total electricity demand increases. The [2] provide a comprehensive description of ice storage systems
utility meets this peak demand through more expensive peaking and propose design guidelines. Many studies have performed field
units. New plants built to meet peak demand are operated at full monitoring of ice storage systems. Authors in [3] describe the ability
capacity only during the short peak periods and remain idle most of an ice storage system to shift a small-sized office building’s peak
of the time. With growing economy the peak demand is constantly cooling demand while maintaining the building’s cooling require-
increasing. Deteriorating load factors, increased use of more inef- ments in California. Authors in [4] analyze the energy performance
ficient and polluting peaking units are the aftermaths of growth of a partial ice storage system operated in a 4-story office building in
in peak demand challenging energy system efficiency and grid San-Ramon, and a 24-story high rise office building in San Francisco.
The results show power savings during peak periods as compared
to three types of modeled conventional cooling systems. Ice stor-
age systems have been modeled and analyzed by several studies.
Abbreviations: ARI, Air-Conditioning and Refrigeration Institute; ASHRAE,
American Society of Heating, Refrigeration, and Air-Conditioning Engineers; COP,
Authors in [5] compare the total energy consumption (including
coefficient of performance; DOE, Department of Energy; DRQAT, Demand Response compressors, pumps and condensing fans) of modeled full and par-
Quick Assessment Tool; LBNL, Lawrence Berkeley National Laboratory; NREL, tial ice storage systems with a conventional cooling system model.
National Renewable Energy Laboratory; PNNL, Pacific Northwest National Labora- The results show the lowest energy consumption for the full stor-
tory; RH, relative humidity; TMY, typical meteorological year; WWR, window to
age system. The partial storage system’s energy consumption is
wall ratio.
∗ Corresponding author. 18% more than that of the conventional cooling system and 24%
E-mail address: [email protected] (F. Sehar). more than that of the full storage system. Authors in [6] use the

0378-7788/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.enbuild.2012.05.002
256 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262

EnergyPlus ice storage model to simulate baseloading operation Table 2

Large and medium-sized office buildings [19].
of partial ice storage systems for a small building in Arizona. The
results show higher chiller energy consumption for ice storage sys- Large-sized Medium-sized
tems than that for non-storage systems for the design day. Authors office building office building
in [4] point out that few thermal energy storage systems take Area 498,588 sqft 53,628 sqft
advantage of daily climatic variations and operating conditions to No. of floors 12 3
optimize system charging and discharging strategies. Authors in Floor to floor height 13 ft 13 ft
WWR 38% 33%
[7] investigate cost savings for different ice storage system’s control
Length 240 ft 164 ft
strategies, different combinations of chiller types, building type and Width 173 ft 109 ft
weather conditions. Authors in [8,9] investigate the prospects of Total no. of occupants 2397 268
using cold thermal storage for office buildings in hot climate. Most Interior lighting power density 1 W/sqft 1 W/sqft
of the studies are based on searching for optimal control strategy Electric plug load density 1 W/sqft 1 W/sqft
to maximize an objective function which can be net energy sav-
ing, cost, comfort condition or some combination of these [10–14].
Authors in [15] model an ice-on-coil internal melt storage tank and setpoints and temperature setpoints for chilled water, condenser
analyze its charging and discharging processes. water and supply air. Some of DRQAT’s output results exploited
No literature work has been found investigating the impacts of in this research include the daily and monthly chiller power con-
ice storage systems on the chiller energy consumption for large sumption for conventional and ice storage cooling systems and the
and medium-sized office buildings in diverse climate zones. This daily ice charge and discharge rates for the ice storage systems.
research paper bridges this knowledge gap by modeling and sim- DRQAT models office buildings for California climate zones and
ulating large and medium-sized office buildings in diverse climate the building construction standards are in compliance with Cali-
zones. The research analyzes the chiller energy consumption for fornia Title 24 construction standard. In order to model buildings
conventional non-storage and ice storage cooling systems for these in selected cities the weather and building construction files in
office buildings. The paper also quantifies peak energy savings as DRQAT have been modified. The weather data for the selected cities
a result of ice storage systems. The ice storage system’s operating as shown in Table 1 has been obtained from NREL TMY3 dataset
and control strategies investigated in this paper include full storage which is available for download as the EnergyPlus weather format
and partial storage with storage priority and chiller priority. from [17]. The large and medium-sized office building construction
This paper is organized as follows. Section 2 discusses the large model details are based on the reference office buildings devel-
and medium-sized office building models with conventional cool- oped by a study conducted by DOE’s Building Technologies Program
ing systems. Section 3 discusses the large and medium-sized office along with PNNL, LBNL and NREL [18]. In this study, 16 types of ref-
building models with ice storage systems. Section 4 discusses the erence commercial buildings have been developed covering 70% of
simulation results. Section 5 summarizes the research findings. US commercial buildings in all US climate zones. The EnergyPlus
input files for the reference office buildings are available in [19].
2. Large and medium-sized office building models with Along with the building construction details the total number of
conventional cooling systems occupants, lighting and electric plug intensities are also based on
reference large and medium-sized office buildings.
This section describes the modeling of large and medium-sized The large and medium-sized office building’s occupancy, light-
office buildings with conventional non-storage cooling systems in ing, electric plug loads, HVAC and cooling set point schedules
selected US cities representing different climate zones. Table 1 are based on the information provided in [20]. These schedules
shows the list of selected cities. have been developed through a combination of industry vali-
The Demand Response Quick Assessment Tool (DRQAT), ver- dated assumptions, reference office buildings and PNNL study
sion 4.0.0, has been used for modeling and simulating the large for medium-sized office buildings. Large and medium-sized office
and medium-sized office buildings in these cities. DRQAT, devel- buildings have been modeled with hydronic and air-cooled systems
oped by LBNL’s Demand Response Research Center, is based on the respectively. The sizing of these conventional cooling systems is
features and capabilities of EnergyPlus and is downloadable from done by EnergyPlus with a global sizing factor of 1.2. The air sup-
[16]. EnergyPlus takes into account the entire building operation plied to the office space is at a temperature of 55 ◦ F. For the hydronic
and design conditions including internal loads and building thermal system the chilled water and condenser water temperature set-
mass. DRQAT unlike EnergyPlus provides a graphical user interface points are 45 ◦ F and 85 ◦ F respectively. These are in compliance with
making it easier to interact with. It requires simple building input ARI 550/590-2003 rating conditions for water chillers.
parameters like the number of floors, the length and width of each Table 2 summarizes general characteristics of both large and
floor, the window-to-wall ratio (WWR), the building location and medium-sized office buildings modeled in DRQAT.
orientation, the number of occupants and densities of internal loads
(lighting and plug loads). DRQAT also allows users to model HVAC 3. Large and medium-sized office building models with ice
and internal loads schedules such as hourly schedules for internal storage systems
loads, operating schedules of the HVAC system, zone temperature
3.1. Overview of ice storage
Table 1
Selected cities representing diverseclimate zones. An ice storage system works by making ice during off-peak hours
to serve part or the entire on-peak cooling requirement. The latent
City Climate zone
heat of water of 144 Btu/lb implies that the melting or freezing of
Miami, FL Hot and humid one pound of ice at 32 ◦ F absorbs or releases 144 Btu of heat energy.
Las Vegas, NV Hot and dry
Baltimore, MD Mild and humid
Ice storage systems are operated either as full or partial storage.
Seattle, WA Marine Full storage unlike partial storage shifts the entire peak cooling
Chicago, IL Cold and humid load to off-peak periods. Partial storage uses chiller and storage
Helena, MT Cold and dry priority strategies to divide the load between the chiller and the
Duluth, MN Very cold
storage. For the chiller priority strategy, the storage meets the
 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262 257

cooling load during peak-hours only when the load exceeds the 2,500
chiller capacity. For the storage priority strategy, the chiller meets
the cooling load during peak-hours only when the load exceeds 2,000
the total storage capacity. This paper studies the ice storage sys-

kW
tems with three different operating strategies, namely full storage, 1,500
partial storage with storage priority and partial storage with chiller
priority. 1,000

3.2. Ice storage model sizing 500

The ice storage model available in DRQAT is an internal melt 0

system with the chiller upstream the ice tank. Ice storage systems
Fig. 1. Miami large-sized office building cooling load profile on design day, July 21st.
include ice harvesting, ice-on-coil internal melt, ice-on-coil exter-
nal melt and encapsulated ice. Ice-on-coil internal melt systems are
Table 3
most commonly used in commercial buildings. In an internal melt Large and medium-sized office buildings in Miami: size of ice storage systems.
system, ice cylinders are formed around submerged coils by cir-
Large-sized Medium-sized
culating cool glycol solution through them. These ice cylinders are
office building office building
melted from the inside by circulating the warm glycol, returning
from the building’s cooling system, through the coils. Total integrated (kWh) 31,210 3,431
cooling load
This model is based on EnergyPlus ice storage model which has Chiller (kW) 2040 224
Full storage
been tested by [6]. DRQAT requires the user to define the capacities Storage (GJ) 112 12
of the chiller and the ice tank. Based on the peak and off-peak peri-
Partial storage: storage Chiller (kW) 1576 173
ods, users can define the operational temperature setpoints of the priority Storage (GJ) 56 6
chiller and the ice storage system, providing them with the ability
Partial storage: chiller Chiller (kW) 1100 130
to implement a number of charging and discharging control strate-
priority Storage (GJ) 25 2.5
gies. Outputs for the ice storage systems analyzed in the research
provided by DRQAT include the chiller power consumption and ice
charge/discharge rates. medium-sized office buildings modeled in Miami. Chiller and ice
Ice storage systems for the large and medium-sized office build- storage capacities have been calculated using cooling load profiles
ings have been sized using ASHRAE cooling design day conditions for Miami and Eqs. (1) and (2).
for the specific cities based on 0.4% annual percentiles. The total
integrated cooling load, obtained from the design day cooling load 4. Simulation results and discussions
profile for the office buildings is used for sizing the ice storage sys-
tems for full and partial storage using equations described in [2]. The 24-h design day and monthly chiller energy consumptions
These equations are presented below. for conventional and ice storage cooling systems have
Total integrated cooling load (kWh)
Chiller capacity = (1)
day time direct cooling (h) + night time charging (h) × capacity factor

Storage capacity = night time charging(h) × capacity factor

been analyzed for large and medium-sized office buildings in
(2) diverse climate zones. The design day peak-hours (noon- 6 p.m.)
energy savings as a result of ice storage systems have also been
Since the chiller for the ice storage system has to cool the glycol
quantified.
solution to lower temperatures during a charge cycle, unlike 45 ◦ F
for the conventional cooling system, the chiller capacity is derated
4.1. Large-sized office buildings
60–70% of nominal capacity [2]. The capacity factor selected in this
study for sizing the ice storage systems is 70%. During the charging
This section presents simulation results and discussions for
period, the coolant temperatures range from 22 ◦ F to 26 ◦ F [2] peak
large-sized office buildings in diverse climate zones. Large-sized
periods, different discharge modes can be implemented. Full stor-
office building’s hydronic cooling system has a water-cooled con-
age, storage priority and chiller priority are implemented by setting
denser with a cooling tower which is a latent heat exchanger. The
chiller outlet temperatures at 55 ◦ F, 50 ◦ F and 45 ◦ F respectively.
evaporative cooling of condenser water depends on the outside air’s

3.3. Office buildings cooling load profiles for ice storage systems
250
DRQAT provides the design day cooling load profiles for the
modeled large and medium-sized office buildings in diverse cli- 200
kW

mate zones. Figs. 1 and 2 show the cooling load profiles for the large
and medium-sized office buildings modeled in Miami respectively. 150
The cooling load for the modeled large and medium-sized office
buildings increases as internal loads increase during the occupied 100
period and outside weather conditions are extreme.
As the modeled large and medium-sized office buildings in 50
diverse climate zones have the same occupancy, lighting and elec-
tric plug schedules and intensities, their internal loads are similar. 0
However outside weather conditions affect the cooling load. Table 3 Fig. 2. Miami medium-sized office building: cooling load profile on design day, July
shows capacities of full and partial ice storage systems for large and 21st.
258 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262

350 make ice and also during day hours to provide direct cooling to the
building. For the full storage system, shown in Fig. 4 the entire peak
300
load is met by storage discharge and the chiller remains nonopera-
250 tional during this period. For the storage priority system, shown in
kW

Fig. 5, the storage discharge remains fairly constant during the peak
200
period and the chiller meets the excess cooling load. For the chiller
150 priority system, shown in Fig. 6, the chiller power consumption also
remains fairly constant during the entire day and any excess load
100
during the peak period is met by storage discharge.
50 The charge rate is initially high when water temperatures are
falling and decreases when ice cylinders start overlapping as heat
0 transfer is restricted. The discharge rate increases gradually as the
Chiller Power Consumpon ice closest to the coil melts first. Heat exchange rate increases
as the ice cylinders break into pieces. As the water temperature
Fig. 3. Miami large-sized office building: chiller power consumption for non storage
starts to rise the discharge rate decreases. Figs. 4–6 exhibit this
system on design day, July 21st.
charge/discharge phenomenon. During the ice making mode the
evaporator temperatures are low thereby reducing the chiller COP
2,000 although this is compensated to some extent by lower night time
temperatures. For the chiller priority system, the chiller COP when
providing direct cooling is slightly higher than the COP during the
kW

1,500
ice making period.
Table 4 shows the 24-h design day and peak-hours (noon to
1,000
6 p.m.) chiller energy consumption for both hydronic and ice stor-
age systems for large office buildings.
500 It can be seen that although the chiller energy consumption for
the 24-h design day is higher for ice storage systems due to their
0 day and night operation but they are able to achieve peak energy
savings.
Chiller Power Consumpon Ice Cooling/Discharge Rate The full storage system in Miami, for example consumes 73%
Ice Charge Rate more chiller energy than the hydronic cooling system. However,
during peak hours the chiller operation is completely eliminated.
Fig. 4. Miami large-sized office building: chiller power consumption and ice
charge/discharge rate for full storage system on design day, July 21st.
For the storage priority system in Miami, although the peak demand
(between noon and 6 p.m.) cannot be shifted entirely, results show
slightly higher chiller energy consumption than that of the full
relative humidity (RH). High RH lowers the evaporative cooling of storage system. This is due to the operation of the chiller during
condenser water, which in turn reduces the chiller coefficient of peak-hours when the outside RH and temperatures are high. For
performance (COP). The ice storage system also has a water-cooled the chiller priority system in Miami, the chiller energy consump-
condenser hence the cooling of condenser water is affected by RH tion is less than that of both the full storage and storage priority
and ambient temperatures. systems. However, the peak demand shifting is almost insignificant
Fig. 3 shows the chiller power consumption in kW of the when compared with the conventional hydronic cooling system.
hydronic cooling system for a large-sized office building in Miami. It is also observed that for Seattle, the chiller priority sys-
The chiller remains nonoperational during the unoccupied period tem achieves more peak energy savings, 39% as compared to the
and turns on to maintain the setup temperature. The chiller con- hydronic cooling system, than in the other climate zones. The chiller
sumes more energy as the cooling load increases during the priority system in Helena achieves the next highest peak energy
occupied period due to the increase in internal loads and outside savings, 26%. This is followed by Duluth at 15%.
weather conditions. Figs. 4–6 show the chiller power consumption Figs. 7–10 analyze the monthly chiller energy consumption for
and ice charge and discharge rates in kW for the full and partial the hydronic and ice storage systems in the seven different cities
ice storage systems for a large-sized office building in Miami. For discussed in Table 1. It can be seen that cities with high RH and
ice storage systems, the ice chiller operates during night hours to ambient temperatures have higher cooling loads; and high cooling

1,000 600

800 500
kWh
kW

400
600
300
400
200
200 100

0 0

Chiller Power Consumpon Ice Cooling/Discharge Rate Chiller Power Consumpon Ice Cooling/Discharge Rate
Ice Charge Rate Ice Charge Rate

Fig. 5. Miami large-sized office building: chiller power consumption and ice Fig. 6. Miami large-sized office building: chiller power consumption and ice
charge/discharge rate for storage priority on design day, July 21st. charge/discharge rate for chiller priority on design day, July 21st.
 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262 259

Table 4
Large-sized office buildings: design day and peak hours chiller energy consumption for non-storage and ice storage systems in diverse climate zones.

Hydronic cooling system Full storage Partial storage: storage priority Partial storage: chiller priority
(kWh) (kWh) (kWh) (kWh)

Design day (24-h) chiller energy consumption

Miami 4267 7391 7530 6530
Las Vegas 3762 8286 7351 6378
Baltimore 3864 6901 6850 5783
Seattle 3019 4502 4071 3152
Chicago 3378 6804 6371 5124
Helena 3682 7165 6231 4843
Duluth 3498 6618 5816 4954
Peak-hours (noon to 6 p.m.) chiller energy consumption
Miami 1753 0 1085 1714
Las Vegas 1606 0 769 1579
Baltimore 1571 0 894 1527
Seattle 1237 0 282 758
Chicago 1375 0 572 1292
Helena 1587 0 711 1182
Duluth 1466 0 630 1243

160000 300000
140000
250000
120000
200000
100000
kWh

kWh
80000 150000
60000
100000
40000
50000
20000
0 0
Jun Jul Aug Sep Jun Jul Aug Sep

Miami Las Vegas Balmore Seale Chicago Helena Duluth Miami Las Vegas Balmore Seale Chicago Helena Duluth

Fig. 7. Large-sized office building: monthly chiller energy consumption for non Fig. 9. Large-sized office building: monthly chiller energy consumption for partial
storage system in diverse climate zones. storage: storage priority system in diverse climate zones.

loads increase the chiller energy consumption. This is because as small for Miami, Baltimore, Chicago, Seattle and Duluth whereas in
the outside fresh air is brought into the building to maintain the Las Vegas and Helena night temperatures are much lower than day
air quality, if this air has higher RH then the latent cooling load temperatures.
increases. Similarly, if the ambient temperatures are high there will Based on temperatures and RH Miami has the highest chiller
be more heat transfer through the building envelope and windows, energy consumption for the summer season followed by Las
increasing the sensible cooling load. Vegas, Baltimore and Chicago. Seattle has the lowest chiller energy
In this study, Miami has high temperatures and RH during the consumption for the summer season compared to others. Both
summer season. Las Vegas has extremely high temperatures, reach- Helena and Duluth have comparable chiller energy consump-
ing 104 ◦ F during the summer season but has low RH. Baltimore and tion. It is observed that for the cooler months in Seattle (June,
Chicago also have high temperatures and RH during the summer July, September), Helena (September) and Duluth (September) the
season. Seattle, Helena and Duluth have mild summer climate. Seat- chiller energy consumption for the chiller priority system is lower
tle’s warmest month is August unlike other cities which have July as than the hydronic cooling system. The chiller energy consumption
their warmest month. The day and night temperature variations are

300000
350000
250000
300000

250000 200000

200000
kWh

150000
kWh

150000
100000
100000

50000 50000

0 0
Jun Jul Aug Sep Jun Jul Aug Sep

Miami Las Vegas Balmore Seale Chicago Helena Duluth Miami Las Vegas Balmore Seale Chicago Helena Duluth

Fig. 8. Large-sized office building: monthly chiller energy consumption for full stor- Fig. 10. Large-sized office building: monthly chiller energy consumption for partial
age system in diverse climate zones. storage: chiller priority system in diverse climate zones.
260 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262

60 120

50 100
kW

kW
80
40
60
30
40
20
20
10
0
0 Chiller Power Consumpon Ice Cooling/Discharge Rate
Chiller Power Consumpon Ice Charge Rate

Fig. 11. Miami medium-sized office building: chiller power consumption for non Fig. 13. Miami medium-sized office building: chiller power consumption and ice
storage system on design day, July 21st. charge/discharge rate for storage priority on design day, July 21st.

for the storage priority system in Duluth during September is also

lower than the hydronic cooling system. These observations result a medium-sized office building in Miami. The ice storage system’s
due to the improved nighttime chiller operation and storage dis- behavior for medium-sized office buildings follows similar expla-
charge available to meet the peak cooling load. nation as described for large-sized office buildings.
Table 5 shows the 24-h design day and peak-hours (noon to
6 p.m.) chiller energy consumption for both the air-cooled and ice
4.2. Medium-sized office buildings
storage systems for medium-sized office buildings.
The full storage system in Miami, for example consumes 39%
This section presents the analysis of the use of ice storage in
more chiller energy than the air-cooled system but eliminates the
medium-sized office buildings in diverse climate zones. Medium-
chiller operation during peak-hours. The storage priority system in
sized office building’s air-cooled system has air-cooled condenser
Miami, shows slightly higher chiller energy consumption than that
which is a sensible heat exchange device. The magnitude of heat
of the full storage system. Same reasoning follows for this observa-
exchange between the refrigerant in the condenser coil and the
tion as described for large-sized office buildings. The chiller priority
outside air is a function of the temperature difference between the
system for the cities shows considerable peak demand shift when
refrigerant and the outside air’s dry bulb temperature. High outside
compared with conventional air-cooled systems unlike large-sized
dry bulb temperatures can increase the chiller energy consump-
office buildings where peak demand shift was insignificant.
tion. The modeled medium-sized office buildings have steel framed
It is also observed that for Seattle, the chiller priority system
exterior walls. Steel is a good conductor of heat energy, a metal
achieves more peak energy savings, 37% as compared to the air-
building freely transfers heat energy in and out of building. The
cooled system, than in the other climate zones. The chiller priority
fiber insulation used can resist heat gain by conduction and con-
system in Helena achieves the next highest peak energy savings,
vection but has little effect on radiant heat gain. Condensation can
31%. This is followed by Las Vegas and Baltimore at 27% and 26%
also occur when there is considerable temperature difference out-
respectively.
side the building. In humid climate zones moisture can be trapped
Figs. 15–18 analyze the monthly chiller energy consumption
within the fiberglass insulation causing its thermal resistance to
for the air-cooled and ice storage systems. The observations can
drop, as water is a good conductor of heat.
be explained on the same reasoning as described for large-sized
Fig. 11 shows the chiller power consumption of the air-cooled
office buildings. However one peculiarity observed is that in July
system for a medium-sized office building in Miami. Like the
and August the air-cooled, full storage and storage priority systems
hydronic system, the chiller for the air-cooled system remains non-
in Las Vegas have higher chiller energy consumption than Miami.
operational during night hours and turns on to maintain the setup
This high chiller energy consumption in Las Vegas can be due to
temperature. The chiller power consumption increases as internal
extremely high temperatures and steel framed buildings. However
loads increase and outside weather conditions become extreme.
for chiller priority, the chiller energy consumption for Las Vegas is
Figs. 12–14 show the chiller power consumption and ice charge
less than that of Miami. This is because Miami has high RH and the
and discharge rates for the full and partial ice storage systems for

250 80
70
200
60
kW
kW

150 50
40
100 30
20
50
10
0 0

Chiller Power Consumpon Ice Cooling/Discharge Rate Chiller Power Consumpon Ice Cooling/Discharge Rate
Ice Charge Rate Ice Charge Rate

Fig. 12. Miami medium-sized office building: chiller power consumption and ice Fig. 14. Miami medium-sized office building: chiller power consumption and ice
charge/discharge rate for full storage system on design day, July 21st. charge/discharge rate for chiller priority on design day, July 21st.
 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262 261

Table 5
Medium-sized office buildings: design day and peak hours chiller energy consumption for non-storage and ice storage systems in diverse climate zones.

Air- cooled system Full storage Partial storage: storage priority Partial storage: chiller priority
(kWh) (kWh) (kWh) (kWh)

Design day (24-h) chiller energy consumption

Miami 586 812 816 801
Las Vegas 517 974 878 754
Baltimore 554 821 799 716
Seattle 236 486 452 346
Chicago 397 776 725 633
Helena 438 827 727 634
Duluth 361 734 657 558
Peak hours (noon to 6 p.m.) chiller energy consumption
Miami 277 0 110 223
Las Vegas 258 0 83 188
Baltimore 261 0 89 194
Seattle 129 0 13 81
Chicago 179 0 56 163
Helena 222 0 78 154
Duluth 171 0 61 142

18000 30000
16000
25000
14000
12000 20000
10000
kWh

kWh
15000
8000
6000 10000
4000
5000
2000
0 0
Jun Jul Aug Sep Jun Jul Aug Sep
Miami Las Vegas Balmore Seale Chicago Helena Duluth Miami Las Vegas Balmore Seale Chicago Helena Duluth

Fig. 15. Medium-sized office building chiller energy consumption for non-storage Fig. 18. Medium-sized office building chiller energy consumption for partial stor-
system in diverse climate zones. age: chiller priority in diverse climate zones.

40000
35000 continuous operation of the chiller for the chiller priority system
30000 may increase its energy consumption.
25000
kWh

20000
5. Conclusions
15000
10000 The impact of ice storage systems on the chiller energy con-
5000 sumption for large and medium-sized office buildings in diverse
0 climate zones has been investigated.
Jun Jul Aug Sep Research findings indicate that the ice storage systems have
Miami Las Vegas Balmore Seale Chicago Helena Duluth higher chiller energy consumptions than the conventional non-
storage systems due to the day and night chiller operation. By
Fig. 16. Medium-sized office building chiller energy consumption for full storage in discharging ice storage during the peak hours, the ice storage sys-
diverse climate zones. tems are able to achieve peak energy savings by reducing or even
eliminating the chiller operation. The chiller energy consumption
35000 for the ice storage systems reflects climatic conditions.
30000 Climate zones with summers having high temperatures and RH
increase not only the building cooling load but also the chiller
25000
energy consumption by decreasing the cooling of condenser water.
20000 For example, Miami with its high temperatures and RH in the sum-
kWh

15000 mer season show high chiller energy consumption for both large
and medium-sized office buildings. Similarly, Las Vegas with very
10000
high temperatures but low RH in the summer season shows high
5000 chiller energy consumption for both large and medium-sized office
0 buildings. On the contrary climate zones with less extreme sum-
Jun Jul Aug Sep mers have lower chiller energy consumption due to lower building
Miami Las Vegas Balmore Seale Chicago Helena Duluth cooling loads and more cooling of condenser water. For example,
Seattle, Helena and Duluth have mild summer seasons and have
Fig. 17. Medium-sized office building chiller energy consumption for partial stor- low chiller energy consumptions for both large and medium-sized
age: storage priority in diverse climate zones.
office buildings.
262 F. Sehar et al. / Energy and Buildings 51 (2012) 255–262

Cost benefit analysis for ice storage systems is beyond the scope [15] Y. Zhu, Y. Zhang, Modeling of thermal processes for internal melt ice-on-coil
of this research paper. However results from this paper will provide tank including ice–water density difference, Energy and Buildings 33 (4) (2001)
363–370.
the necessary energy consumption savings information required [16] PIER Demand Response Research Center, retrieved June 01, 2011,
for such cost benefit analyses. The analysis presented here provides http://poet.lbl.gov/drrc/tools-drqat-download.html.
guidelines for utilities and building owners to determine potential [17] EnerguPlus Energy Simulation Software, updated March 17, 2011,
retrieved May 01, 2011, http://apps1.eere.energy.gov/buildings/energyplus/
benefits of ice storage systems for different climate zones and ice cfm/weather data3.cfm/region=4 north and central america wmo region 4/
storage operating strategies. country=1 usa/cname=USA.
[18] M. Deru, K.Field, D. Studer, K. Benne, B. Griffith, P. Torcellini, U.S. Department of
Energy Commercial Reference Building Models of the National Building Stock,
References NREL/TP-5500-46861, 2011.
[19] US Department of Energy, updated September 30, 2010, retrieved July 01, 2011,
[1] I. Dincer, M.A. Rosen, Thermal Energy Storage Systems and Applications, 2nd http://www.eere.energy.gov/.
edition, John Wiley, Hoboken, N.J., 2011. [20] M. Leach, L. Chad, A. Hirsch, S. Pless, P. Torcellini, Technical Support Document:
[2] C.E. Dorgan, J.S. Elleson, Design Guide for Cool Thermal Storage, American Soci- Strategies for 50% Energy Savings in large office Buildings, NREL/TP-550-49213,
ety of Heating and Air-Conditioning Engineering, Atlanta, 1993. 2010.
[3] D.R. Laybourn, V.A. Baclawski, The benefits of thermal energy storage for cool-
ing commercial buildings, power apparatus and systems, IEEE Transactions Fakeha Sehar (S’11 – IEEE) is pursuing her M.S. degree in the Department of Electri-
PAS–104 (9) (1985) 2356–2360. cal and Computer Engineering at Virginia Polytechnic Institute and State University,
[4] H. Akbari, O. Sezgen, Case studies of thermal energy storage (TES) systems: eval- VA, USA. She received her B.E. degree in 2006 in Electronics Engineering from NED
uation and verification of system performance, LBL-30852, Lawrence Berkeley University of Engineering and Technology, Karachi, Pakistan. Her fields of interest
Laboratory, University of California, Energy and Environment Division, Berke- include sustainable building design, demand response, smart grid and renewable
ley, CA, 1992. energy systems.
[5] C. Chatchawan, W.W.S. Charters, L. Aye, An ice thermal storage computer
model, Applied Thermal Engineering 21 (17) (2001) 1769–1778. Saifur Rahman (S’75, M’78, SM’83, F’98 – IEEE) is the director of the Advanced
[6] P. Ihm, M. Krarti, G.P. Henze, Development of a thermal energy storage model Research Institute at Virginia Tech where he is the Joseph Loring Professor of elec-
for EnergyPlus, Energy & Buildings 36 (2004) 807–814. trical and computer engineering. He also directs the Center for Energy and the Global
[7] G.P. Henze, M. Krarti, M.J. Brandemuehl, Guidelines for improved performance Environment at the university. In 2011 he is serving as the vice president for New
of ice storage systems, Energy & Buildings 35 (2003) 111–127. Initiatives and Outreach of the IEEE Power & Energy Society and a member of its
[8] S.M. Hasnain, S.H. Alawaji, A.M. Al-Ibrahim, M.S. Smiai, Prospects of cool ther- Governing Board. He is a member-at-large of the IEEE-USA Energy Policy Commit-
mal storage utilization in Saudi Arabia, Energy Conversion & Management 41 tee. Professor Rahman is currently the chair of the US National Science Foundation
(2000) 1829–1839. Advisory Committee for International Science and Engineering. He has served as a
[9] S.M. Hasnain, N.M. Alabbadi, Need for thermal energy storage in Saudi Arabia, program director in engineering at NSF between 1996 and 1999. In 2006 he served
Applied Energy 65 (2000) 153–164. as the vice president of the IEEE Publications Board, and a member of the IEEE Board
[10] Ali E.H. Hajiah, Development and implementation of an optimal controller of of Governors. He is a distinguished lecturer of IEEE PES, and has published over 300
a central cooling plant using ice storage system and building thermal mass, papers on conventional and renewable energy systems, load forecasting, uncertainty
Department of Civil, Environmental and Architectural Engineering, University evaluation and infrastructure planning.
of Colorado: Boulder, 2000. Manisa Pipattanasomporn (S’01, M’06 – IEEE) joined Virginia Tech’s Department
[11] H.-J. Chen, D.W.P. Wang, S.-l. Chen, Optimization of an ice-storage air condition- of Electrical and Computer Engineering as an assistant professor in 2006. She serves
ing system using dynamic programming method, Applied Thermal Engineering as one of the principal investigators (PIs) of multiple research grants from the U.S.
25 (2–3) (2005) 461–472. National Science Foundation, the U.S. Department of Defense and the U.S. Depart-
[12] M. Krarti, M.J. Brandemuehl, G.P. Henze, Evaluation of optimal control for ice ment of Energy, on research topics related to smart grid, microgrid, energy efficiency,
storage systems, ASHRAE Report, 1995. load control, renewable energy and electric vehicles. She received her Ph.D. in elec-
[13] D.D. Massie, Optimization of a building’s cooling plant for operating cost trical engineering from Virginia Tech in 2004, the M.S. degree in Energy Economics
and energy use, International Journal of Thermal Sciences 41 (12) (2002) and Planning from Asian Institute of Technology (AIT), Thailand in 2001 and a
1121–1129, Fuel and Energy Abstracts 44 (4) (2003) 254. B.S. degree from the Electrical Engineering Department, Chulalongkorn University,
[14] M. Kintner-Meyer, A.F. Emery, Optimal Control of an HVAC system using cold Thailand in 1999. Her research interests include renewable energy systems, energy
storage and building thermal capacitance, Energy and Buildings 23 (1) (1995) efficiency, distributed energy resources, and the smart grid.
19–31.