SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 1

Development of a Smart Residential Load Simulator

for Energy Management in Smart Grids
Edris Pouresmaeil, Juan Miguel Gonzalez, Member, IEEE, Claudio A. Canizares, Fellow, IEEE, and Kankar
Bhattacharya, Senior Member, IEEE

Abstract—This paper describes the development of a freeware I. I NTRODUCTION

Smart Residential Load Simulator to facilitate the study of
energy management systems in Smart Grids. The proposed tool MART grids coupled with renewable energy resources
is based on Matlab-Simulink-GUIDE toolboxes and provides a
complete set of user-friendly graphical interfaces to properly
S can yield significant economic and environmental benefits.
The Smart grid’s ability to improve safety and efficiency, make
model and study smart thermostats, air conditioners, furnaces,
water heaters, refrigerators, stoves, dish washers, cloth washers, better use of existing assets, enhance reliability and power
dryers, lights, and poolpumps, as well as wind, solar, and battery quality, reduce dependence on imported energy, and minimize
sources of power generation in residential houses. The impact of environmental impacts is a market force that has substantial
different variables such as ambient temperature, solar radiation, economic value. These grids are growing fast, but if this is
and household activity levels which considerably contribute to
energy consumption are considered. The proposed simulator
to be sustained, their value must become more clear to all
allows to model the way appliances consume power and helps to stakeholders, especially residential consumers. The latter are
understand how these contribute to peak demand providing in- an important part of electricity demand, since for example,
dividual and total energy consumption and costs. In addition, the the residential sector accounted for 25% of the peak demand
value and impact of generated power by residential power sources and almost 30% of the electrical energy demand in Ontario,
can be determined for 24h horizon. This freeware platform can
be a useful tool for researchers and educators to validate and
Canada in 2005 and 2008 [1], [2]. Also, residential energy
demonstrate models for energy management and optimization, consumption in the US was 22% of the total consumed energy
and can also be used by residential costumers to model, and in 2010 [3], and similar values were reported for the European
understand energy consumption profiles in households. Some Union in 2009 [4].
simulation results are presented to demonstrate the utilization Space heating/cooling systems, water heaters, refrigerators,
and performance of the proposed simulator.
dishwashers, cloth washers, dryers, lighting, and cooking
Index Terms—Smart grids, house appliances, smart thermo- ranges are the most common appliances in the residential sec-
stat, energy management. tor [1]-[4]. Heating, ventilation, and air conditioning (HVAC)
ABBREVIATIONS and water heaters are major energy consumption devices.
AC Air Conditioner Therefore, controlling the residential end-use electricity de-
APACHE Applications Program for Air-Conditioning mand can have a significant impact on reducing the peak
demand and optimize energy consumption, which can be
and Heating Engineers
accomplished in smart or intelligent homes with automation
BTU British Thermal Unit systems to control residential loads [5]-[6].
CFL Compact Fluorescent Lights Several studies have been reported in the literature regarding
CHVAC Commercial Heating,Ventilation and Air- the prediction of load-shape and optimization methods for
Conditioning energy management, since some appliances can be easily
CW Clothwasher scheduled to reduce energy cost and consumption without
affecting customer comfort. For instance, a model to minimize
DW Dishwasher
the peak load by scheduling pool pumps, air conditioners
HVAC Heating, Ventilation and Air Conditioning (AC) and water heaters (WH) is proposed in [7]; a mixed
SRLS Smart Residential Load Simulator integer linear programming model is developed to minimize
TOU Time of Use the energy cost and maximize costumers comfort while taking
WH Water Heater into account the influence of price signals on the household.
HT Furnace Some projects focus on scheduling the start-up of HVAC
and/or water heater by making use of wireless thermostat
SOC State-Of-Charge
technology to optimize cost and thermal comfort, as in the
case of [8].
This work is supported by a Canadian Natural Sciences and Engineering
Research Council (NSERC) Strategic Grant. References [9]-[15] explore different ways of creating
E. Pouresmaeil, C.A. Canizares, and K. Bhattacharaya are with the Depart- appliance-level load models based on statistical data to predict
ment of Electrical and Computer Engineering, university of waterloo, Water- load-shape demand for load management purposes. Some
loo, N2L 3G1, Canada (email: epouresm, ccanizar, kankar@uwaterloo.ca).
J.M. Gonzalez is with CINVESTAV, Guadalajara, Jalisco, CP 45010, simulators have been developed to model HVAC systems and
Mexico (email: juangol7906@hotmail.com). buildings, such as the EnergyPlus software which models
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 2

thermal energy in building [16]. The CHVAC software calcu- a classical thermal model [19],[20]. Storage tank water heaters
lates the maximum heating and cooling loads for commercial are the most common types used in North America; therefore,
buildings [17]. The Applications Program for Air-Conditioning electric and gas storage tank water heaters are modeled in the
and Heating Engineers (APACHE) is a graphical user interface SRLS.
to analyse thermal performance and energy use of buildings Figure 3 shows the graphical interface of the WH in the
[18]. None of the existing modeling tools take into account SRLS. The inlet water and ambient temperatures around the
other appliances and some are not easy to use. Hence, there is a tank, capacity of the WH, and its efficiency are considered
need for user-friendly simulator to understand how appliances as inputs. The power consumption is reported in W when an
interact with each other with respect to energy consumption, electric WH is chosen, and in BTU for a gas WH. In both
as well as facilitate the study and application of mathematical cases, typical values are provided as default, corresponding
models for home energy management systems. to values applicable in southern Ontario, Canada, for inlet
This paper presents the development of a smart residential water and ambient temperatures. Generally, the efficiency of
load simulator with a user-friendly graphical interface, which electric WHs are in the range of 85-94%, while for gas
can simulate optimal on/off decisions of residential appliances WHs is 50-65%. More details and information regarding the
to validate residential optimization models for a 24h horizon. WH mathematical model are provided by pressing the “Help”
Local power generation such as roof-top-solar, battery energy button.
storage, and wind turbines are considered in this simulator, so Figure 4 shows the circuit used to model the WH, which
that the benefits of these resources can be also studied. comprises the mass of water (m), specific heat of water (Cp ),
The rest of the paper is structured as follows: Section characteristics of fiber glass (CW , U A), gas or electric rated
III presents a general description of the proposed simulator, power (Qe− g ), and the efficiency (η) [19]. The following
together with the associated graphical interfaces, and explains equation represents the energy flow in the WH that is used
the main appliances’ interfaces and models. An example is pre- to implement the model:
sented to demonstrate the functionality of proposed simulator
in Section IV. Finally, the main conclusions and contributions
of the presented work are provided in Section V. dTw mCp UA U A + mCp
= Tinlet + Tamb − + Qeg η (1)
dt Cw Cw Cw
II. S MART R ESIDENTIAL L OAD S IMULATOR (SRLS)
where Tw is the temperature of the tanks’ wall, Tinlet is the
The developed tool which is available at
inlet water temperature, and Tamb is the ambient temperature
www.power.uwaterloo.ca is a new Matlab-based simulator
around the tank. The procedure to calculate the hot water usage
that represents most of the important residential loads and
is explained in detail in [21], which depends on the number
power sources. The toolbox is provided with a complete
of family members.
graphical interface as shown in Fig.1. Factors such as solar
radiation and ambient temperature that play important roles
on the energy consumption of a household are considered as
user-defined inputs to the SRLS. Other inputs are electricity B. Household
tariffs by season and time-block rates (off-peak, mid-peak,
The material properties of buildings influence the thermal
and on-peak) to represent Time of Use (TOU) tariffs. The
performance and their energy consumption patterns. The walls,
user can also define real time prices (RTP). All and each
floor, roof and windows have central thermal conductivity, and
of the appliances shown in Fig.1 are modeled and can be
allow circulation of warm/cold air in the house. The energy
simulated individually or as a group.
consumption depends on the house characteristics, specifically
Figure 2 shows the interface for plotting the simulation
on its geometry. Therefore, in SRLS, the house geometry is
results, where consumed and generated power by the appli-
defined by the size and the numbers of rooms, which are
ance and sources is illustrated together with the levels and
assumed to be from 1 to 4, modelled using the average of
costs of consumed and generated energy. In addition, the
length, width and height of walls and windows. The thermostat
user can select each appliance and resource individually to
is assumed to be placed in one of the rooms. Figure 5 shows
plot its energy consumption/generation profile.The charge and
the graphical interface to represent the house where the user
discharge profiles of battery storage can be also depicted.
inputs the required house, profile information.
Moreover, the interface provides consumption and generation
tables where the cost of consumed energy by appliances and Figure 6 depicts the circuit model used to represent a single
sources during off, middle, and on-peaks periods are detailed. room, which considers the outside environment (Tamb ), the
Finally, gas consumption and its costs can be also illustrated thermal characteristics of the room (thermal resistance of
by the interface. The models of the appliances and energy walls Rw and windows Rc , and thermal capacitance of the
sources considered in the simulator and explained next. wall Cw and indoor air Ci ), and the AC or furnace system
which are represented by the Qac− ht thermal source. Using
this model, the room’s temperature, the power consumption
A. Water Heater (WH) in the room, and the corresponding cost of consumed energy
The WH is a cylindrical tank enclosed by insulation and can be calculated. The following differential equations can be
covered with a metal sheet, which can be simulated by using obtained from Fig. 6 [22], [23]:
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 3

Fig. 1. Graphical interface of Smart Residential Load Simulator.

Fig. 2. Graphical interface for plotting of simulation results.

C. Air Conditioner (AC)

The AC equipment is often specified by its cooling capacity

dTw Qs Tout Tin 2Tw in terms of BTU. This capacity is the amount of energy used
= + + −
dt Cw
Rw Cw Rw C w Rw C w by the equipment to remove heat from the air, and regulate
Qin − Qac− ht S (t)
 
dTin Tin 1 1 the temperature and humidity in a room or the entire house.
= − +
dt Cin Cin Rw Rc There are two types of AC systems: window and central AC.
Tw A typical window AC has a capacity of around 6000-18000
− (2) BTU. A central AC with split configuration uses ducts or
Rw Cin
pipes to distribute cool air to one or more rooms, and its
typical capacity is around 9000-60000 BTU. Figure 7 shows
where S is a binary variable representing the ON (1) or OFF the graphical interface of the AC in the SRLS, where the user
(0) state of the AC/HT. can select the capacity of the equipment.
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 4

Air conditioner

Water_Heater

9000 Capacity (BTU)

Gas Type of water heater
230 Voltage (Volts)
10 Temperature of inlet water (C)

4.5 Current (Ampers)

20 Temperature of ambient (C)

Rated Power in (Watts if 880 Power (Watts)

40000
electric or BTU if gas)

10 Energy Efficiency Ratio (EER)

184 Capacity in liters

0.62 Efficiency
Continue Help

Set o
Point 55 C Continue Help

Fig. 7. AC graphical interface.

Fig. 3. Graphical interface for water heater. Qout Qout

EER AFUE
Win
Qht
Qin Qin
Tinlet m Cp (a) (b)
Tw UA
Ta
Cw Fig. 8. Carnot machine representation of (a) AC and (b) HT.
Qe_gh
The modeling of the AC is represented schematically by the
heat flow diagram in Fig. 8 (a). The Energy Efficiency Ratio
(EER) denotes the amount of cooling effect provided by the
Fig. 4. Circuit model for water heater.
AC as follows:
−Qin Qin
EER = 3.412 = (3)
Household Win Qin − Qout
Four rooms can be simulated, set the number of room you want to simulate and fill each box. The whole house could be considered
as one room, place the average of length, width and height of walls as well as windows. where Qout is the required energy used to extract the heat
Qin from the rooms, and the electrical input Win represents
4 # of rooms (1-4)
the energy required to do this work.
Thermostat is placed:

Inside room 1

D. Furnace (HT)
ROOM 1 (all parameters in meters) ROOM 2 (all parameters in meters) Central gas furnaces are normally used in households to
7 Length of room #1 Windows? 3.5 Length of room #2 Windows?

6 Width of room #1
Yes
4 Total windows length 4 Width of room #2
Yes
1 Total windows length
inject hot air into the rooms. The most common type is a
2 Height of room #1 1.5 Total windows width 2 Height of room #2 1 Total windows width natural gas fired furnace inside an enclosed metal casing,
ROOM 3 (all parameters in meters) ROOM 4 (all parameters in meters) which injects and distributes heated air in the house. The
4 Length of room #3 Yes Windows? 3 Length of room #4 Yes Windows?
graphical interface of the furnace is shown in Fig. 9, where
4 Width of room #3 1 Total windows length 3 Width of room #4 1 Total windows length

2 Height of room #3 1 Total windows width 2 Height of room #4 1 Total windows width only the capacity and Annual Fuel Utilization Efficiency
Continue Help (AFUE) values are needed as inputs.
The heat flow diagram of the heating system is depicted in
Fig. 8 (b), where the efficiency is known by the furnace AFUE
Fig. 5. Graphical household interface. rating. The following equation represents the thermal model
of the furnace.
Qin Qin
AF U E = 3.412 = (4)
Rc Qht Qin − Qout
where Qht represents the capacity of the furnace and Qin
Rw Tw Rw Tin
represents the heat inside the house.
S(t)

Tamb Qs Qac_ht
E. Smart Thermostats
Cw Ci
Qin
Programmable thermostats are used in most households with
central AC and/or HT. Such thermostat is designed to adjust
the temperature according to user preferences at different
times of the day, and helps regulate the home temperature in
Fig. 6. Thermal circuit model of a single room. both summer and winter. Therefore, the thermostat can be set
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 5

Furnace Stove

Type

Electric
Size of Size of
burner (in) burner (in)
6 6
Times of use

40000 Capacity (BTU)

Morning 7.30 Hour of the day when switched on?
Size of Size of 40 Duration of use in (min)
burner (in) burner (in)
Annual Fuel Utilization 6 14.30 Hour of the day when switched on?
10 Efficiency (AFUE) 6 Noon
90 Duration of use in (min)

20.30 Hour of the day when switched on?

Night
Continue Help 50 Duration of use in (min)

0 0
% Intensity Continue Help
0 0

Fig. 12. Graphical interface for stove.

Fig. 9. Graphical interface for furnace.

Lights_Parameters

P1 P2 P3 P4 INCANDESCENT LIGHTING CFL LIGHTING

23 o C Set Time 6 8 17 22
6
Hour of the day when turns on during
the morning?
6
Hour of the day when turns on during
the morning?
2 Power-on hours on morning 2 Power-on hours on morning
Set to 21 27 23 21
Hour of the day when turns on during Hour of the day when turns on during
0.5 +/- 18 18
+/- 1 2 1 0.5 the night? the night?
4 Power-on hours on night 4 Power-on hours on night

100 Power (Watts) 9 Power (Watts)

On AC Sch off On AC Sch On

0 How many bulbs? 0 How many CFLs?

Help Help

(a) (b) FLUORESCENT TUBE LIGHTING

Hour of the day when turns on during
6
the morning?
2 Power-on hours on morning
Hour of the day when turns on during
Fig. 10. Graphical interface for (a) conventional and (b) programmable 18
the night?

thermostat. 4 Power-on hours on night

15 Power (Watts)

0 How many fluorescent tubes?

Help
according to the family’s schedule and preferences to regulate Continue

the temperature of the house.

Both conventional and programmable thermostats are con-
sidered in the SRLS. Figure 10(a) illustrates a conventional Fig. 13. Graphical interface for lighting.
thermostat, where the user has to select the desired tempera-
ture. Figure 10(b) depicts a programmable thermostat where G. Lighting
the user can specify four time periods, as well as upper The most common types of lights used in residential houses
and lower temperature set points. Figure 11 illustrates the are the traditional incandescent bulbs, Compact Fluorescent
thermostat model used in this simulator, where Thi and Tlo are Lights (CFL), and fluorescent tubes. Residential houses usu-
the upper and lower temperature limits, respectively, within ally use a mixture of these three types of lights. CFL and
which the thermostat maintains the house temperature. This fluorescent tubes are more expensive, but they have a longer
values are set by the user pressing the +/− button. life and use much less energy, thus resulting in significant
savings in energy and cost.
Figure 13 shows the graphical interface for the lighting
F. Stove system in the SRLS. The number, power rating, and operation
(time and duration of use) of the lights are determined in this
Normally, gas or electricity stoves are used. About 87% of interface, from which their energy consumption can be readily
families in the U.S use electric range-ovens for cooking [3]; calculated.
therefore, only electrical stoves are considered in the SRLS.
Energy consumption in the stove is calculated by multiplying
H. Refrigerator
the consumed power by the duration of use. The graphical
interface of the electrical stove is depicted in Fig. 12, where it The refrigerator is modeled as a thermal system with an
is possible for the user to select the number of heating elements insulation of fiber glass. The corresponding model is similar
and their corresponding heat intensity for three time periods to the room model mentioned earlier; therefore, it can be
in a day. represented using the same circuit model by simply changing
the parameter values [22]. Figure 14 depicts the graphical
interface used to define the refrigerator main characteristics.
Thi
Set point

I. Dryer
t
Gas and electric dryers use large amounts of energy in
Tlo
a household. Electrical dryers are commonly used in North
America, and hence only these are considered in the SRLS.
Fig. 11. Thermostat on/off decisions. An example of the energy consumption pattern of a dryer is
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 6

Refrigerator

P2 P4
127 Voltage (Volts)

2.4 Current (Ampers)

1.7m
P1 P3 P5
350 Power (Watts)

o
5 C Fig. 17. Power consumption cycle for Dishwasher.
0. Set Point
Wi 9m
dth 0.8m
Length
Continue Help
Dishwasher

From Yellow Energy Guide Label

Annual energy consumption

3 # Loads per day Hour switched on?

Fig. 14. Refrigerator graphical interface. Check if is connected to hot water

Minutes of the load Hour switched on?
50

Panel Hour switched on?

Low efficiency
P1 Energy star Continue Help

P2 Fig. 18. Graphical interface for Dishwasher.

60 min 120 min amounts of power P3. The dishes are dried using the electric
resistance element consuming P4 power, and finally, the hot
Fig. 15. Power consumption pattern of dryer. air remaining in the DW, consuming P5 power. According to
[26], about 55% of the energy used by a DW goes to heat
shown in Fig. 15 [25], where power P1 is in the range of the water when connected to a WH, and 65% if cold water
2,000 to 2,500 W during the first period and P2 is 500 W for is used. The time period of power consumption depends on
the next period. In the SRLS, a typical 2,000 W is assumed the efficiency of the DW. The SRLS model fits Fig. 17 to
for the first 60 minutes of use, and 500 W for the remaining the Yellow Energy Guide under standard conditions, and the
period. Figure 16 shows the interface for the dryer, where the specifications provided by the user in the graphical interface
user can select up to three loads per day and the corresponding shown in Fig. 18. Three loads per day, duration and time of
duration of use. use can be entered by users.

K. Clothwasher (CW)
J. Dishwasher (DW)
The CW process is controlled by a step timer or an
The DW represent a small share of residential appliances electronic control device. Electrical energy is used mainly for
energy consumption. However, DWs draw a high power during driving the drum motor and heating up the water, if it is not
short periods of time, which makes them relevant for peak hot enough, in spite of the fact that about 2/3 to 3/4 of the
demand programs [27]. water used is cold water for rinsing [27],[28].
Figure 17 shows the sequence of operations of a typical DW. Figure 19 shows the graphical interface for CW in the
At first the DW fills up with water for about 15 minutes and a SRLS. The number of loads per day, time and duration of
constant power P1 is drawn; then, it provides electric heating, use, water temperature, and type of efficiency can be input
increasing its power to P2 for a time period that depends if by the user. An example of the CW power demand profile is
it is connected to hot water or cold water [1]. After that, shown in Fig. 20, where P1 and P4 powers correspond to the
hot water and detergent are sprayed over the dishes, draining filling and draining of rinse water, and P2 and P3 correspond
and refilling alternatively with rinse water; this consumes an to heating the water. The model developed in the SRLS to
determines this powers from the Yellow Energy Guide and
Dryer
the user defined inputs.

L. Poolpump
1 # Loads per day Considerable amount of energy is needed for heating and
50 Minutes of the load
maintaining water temperature in pools, in addition to the
energy used by the poolpump to circulate and filter the pool
Hour switched on? water. Pool water heating can be solar, gas, or by an electrical
heat pump. In a swimming pool, 76% of electrical energy is
Continue Help used for pumps, 6% for chlorination cells, 14% for electric
heaters, and 4% for timers and controls [29].
A typical poolpump consumption pattern is shown in Fig.
Fig. 16. Graphical interface for dryer. 21. Generally 200-500W single-phase pumps are used for
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 7

Clothwasher Wind PV
Hour Power Hour Power
From Yellow Energy Guide Label 1 2965 1 0
2 4220 2 0
Annual energy consumption 3 2825 3 0
4 4555 4 0
5 3220 5 0
6 3711 6 0
# Loads per day Hour switched on? 7 2430 7 0
3 8 2000 8 0
Check if is connected to hot water 9 2650 9 25.85
10 1700 10 74.18
Minutes of the load Hour switched on? 11 2440 11 195.32
30
12 2250 12 281.63
13 2680 13 453.22
Cold Water temperature? Hour switched on? 14 3100 14 425.31
15 4510 15 556.11
16 5150 16 445.22
17 4550 17 485.13
Panel 18 3200 18 360.17
19 4100 19 253
Low efficiency 20 3600 20 126.11
Energy star 21 4000 21 33.25
Continue Help 22 2600 22 0
23 5220 23 0
24 3600 24 0

Continue Help Continue Help

Fig. 19. Graphical interface for clothwasher. (a) (b)

Battery
Hour SOC(%)
0 35
1 55
P2 2
3
20
70
4 25
5 30
6 45
P3 7
8
60
45
9 80
P1 P4 10
11
85
55
12 92
13 76
14 51
15 31
16 62
17 87
Fig. 20. Power consumption cycle of clothwasher. 18
19
97
75
20 61
21 43
residential swimming pools, with 3 to 8 working hours per 22
23
29
59

day for water filtration, depending on the pool size, pump Continue Help

size, environmental conditions such as outside temperature and (c)

sunshine, water filtration equipment, how often the pool is
used, and other pool manufacturer recommendations. Usually, Fig. 23. Graphical interface for local power generation resources: (a) Wind,
poolpumps are controlled by electro-mechanical or electronic (b) PV, and (c) Battery.
ON/OFF clock timers with start- and end-times manually
selected by users. Figure 22 presents the interface for the user
These power sources are not dispatchable and are available at
to define up to three loads per day, specifying the time and
periods of low demand; therefore, they are typically integrated
duration of use.
with some storage devices, such as batteries, to store the
generated energy for a certain period of time, releasing it
M. Local Generation Resources when demand increases. However, besides being expensive,
Wind and solar photovoltaic (PV) power generation are batteries have limited capacity; thus, if there is a surplus of
considered as local power sources supplying residential loads. energy produced by, for example, a domestic PV system, this
extra energy is sold to the local grid.
Figure 23 depicts the window interfaces for the user for
P1 wind, PV, and battery systems. In Fig. 23(a) and Fig. 23(b)
different power outputs per hour are defined for wind and
P2
PV generations. Figure 23 (c) shows the window interface
for the battery, where the user can select the kWh rating and
SOC hourly profile for the day. The sum of these three power
sources is assumed to supply the load or be injected into the
grid.
Fig. 21. Power consumption cycle of poolpump.

III. R ESULTS AND D ISCUSSION

The AC and gas WH are considered here as an example of
residential loads, and solar PV and a battery are selected as
sources of local power to illustrate the SRLS. Thus, an AC
with 48,000 BTU is used to cool the air in a house comprised
of four rooms, inputting the required information for the rooms
as shown in Fig. 5. Figure 7 illustrates how the user should
input the AC parameters in the simulator. The thermostat is set
at 23o C with a +/− 0.5o C tolerance, as in Fig. 10 (a). Figure
3 shows the information required to model the gas WH.
The simulator takes approximately 20 seconds to solve
Fig. 22. Graphical interface for poolpump. the model equations, generating data for the user to analyze
SUBMITTED TO IEEE TRANSACTIONS ON POWER SYSTEM 8

[4] P. Bertoldi and B. Atanasiu, “Electricity Consumption and Efficiency

Trends in European Union,” JRC Scientific and Technical Reports, 2009.
[5] “The Smart Grid: An Introduction,” U.S. Department of Energy, Office
of Electricity Delivery and Energy Reliability, Jan. 2010.
[6] R. G. Pratt, “Transforming the U.S. electricity system,” in Proc. of IEEE
PES Power Syst. Conf. and Exhibition, New York, Oct. 2004.
[7] C. N. Kurucz, D. Brandt, and S. Sim, “A linear programming model for
reducing system peak through customer load control programs,” IEEE
Trans. Power Systems, vol. 11, no. 4, pp. 1817-1824, Nov. 1996.
[8] “Demand responsive electrical appliance manager,”. [Online]. Available:
http://dr.berkeley.edu/dream/
[9] A. Capasso, “A Bottom-up Approach to Residential Load Modeling,”
IEEE Trans. on Power Systems, vol. 9, no. 2, 1994.
[10] R. Yao, and K. Steemers, “A method of formulating energy load profile
for domestic buildings in the UK,” Energy and Buildings, vol. 37, pp
663-671, 2005.
[11] J.V. Paatero, and P.D. Lund, “A model for generating household elec-
tricity load profiles,” Int. J. Energy Research, vol. 30, pp. 273-290, 2006.
[12] C. Jardine, “Synthesis of High Resolution Domestic Electricity Load
Fig. 24. Water heater details. Profiles,” in Proc. Workshop on Micro-Cogeneration and Applications,
April 29 - May 1, 2008. Ottawa, Canada.
[13] I. Richardson, “A high-resolution domestic building occupancy model
the behavior of the simulated appliances. Figure 2 shows for energy demand Simulations,” Energy and Buildings, vol. 40, pp. 1560-
1566, 2008.
the consumed and generated energy by the loads and local [14] I. Richardson, “Domestic electricity use: a high-resolution energy de-
generation sources, respectively. The WH and AC loads and mand model,” Energy and Buildings, vol. 42, 1878-1887, 2010.
the corresponding total consumed energy are shown along with [15] M.M. Armstrong, “Synthetically Derived Profiles for Representing
Occupant-Driven Electric Loads in Canadian Housing,” J. Building Per-
the battery and the solar PV output. The cost of energy used formance Simulation, vol. 2, 15-30, 2009.
by the AC and WH and the cost of energy saved from local [16] “EnergyPlus Energy Simulation Software”. [Online]. Available:
sources are also illustrated. The defined SOC of the battery http://apps1.eere.energy.gov/buildings/energyplus/
[17] “Chvac - Commercial HVAC Loads”. [Online]. Available:
and the inside house temperature, outside temperature, and http://www.elitesoft.com/web/hvacr/chvacx.html
AC power are also shown in this figure. The Consumption [18] “APACHE”. [Online]. Available: http://httpd.apache.org/
and Generation tables in the figure illustrate the value of [19] K. Elamari, L.A.C. Lopez, and R. Tonkoski, “Using electric water
heaters (EWHs) for power balancing and frequency control in PV-Diesel
consumed electricity and gas, and the generated energy by Hybrid, mini-grids,” in Proc. World Renewable Energy Congress, Sweden,
the local generation, during off-, mid-, and on-peak hours, 2011.
respectively. Finally, Fig. 24 shows the hot water temperature [20] J. Lutz, X. Liu, J. McMahon, C. Dunham, L. Shown, and Q. McCure,
“Modeling patterns of hot water use in households,” Lawrence Berkeley,
and consumption, and water heater power generated by the National Laboratory, 1996.
SRLS. [21] Hourly water heating calculations, “Pacific Gas and
Electric Company,” Tech. Rep., 2002. [Online]. Available:
http://www.energy.ca.gov/title24/2005standards/archive/documents/
IV. C ONCLUSIONS 2002-05-30 workshop/2002-05-17 WTR HEAT CALCS.PDF
[22] A. Molina, A Gabeldon, J.A. Fuentes, and C. Alvarez, “Implementation
A new toolbox based on Matlab-Simulink has been de- and assessment of physically based electricl load models: application
veloped to model residential energy consumption and local to direct load control residential programmes,” IEE Proc-Gener. Tranm.
Distrib., vol. 150, no. 1, Jan. 2003.
generation resources. The main objective of the proposed [23] R. J. Gran, “Numerical Computing with Simulink,” Volume I. SIAM
simulator is to allow the study, demonstrate, and teach energy Philadelphia, 2007
management of residential households. The simulator has been [24] “Central Air Conditioners,”. [Online]. Available:
http://www.furnacecompare.com/air-conditioners
described together with the models and graphical interfaces of [25] R. Stamminger, “Synergy Potential of Smart Appliances,”
the main residential energy consuming appliances and local Report D2.3 of WP 2 from the Smart-A project University
generation, and an example illustrating its performance and of Bonn, March 2009. [Online]. Available: http://www.smart-
a.org/WP2− D− 2− 3− Synergy− Potential− of− Smart− Appliances.pdf
application has been presented. This tool can be useful for [26] D.E. Hoak, D.S. Parker, and A.H. Hermelink, “How energy efficient are
researchers to validate their models for energy management modern dishwashers,” in Proc. of ACEEE, Florida Solar Energy Center,
and optimization, and can also be used by costumers and 2008.
[27] M. Eastment and R. Hendron, “Method for Evaluating Energy Use of
educators to understand and explain residential energy demand Dishwashers, Clothes Washers, and Clothes Dryers,” National Renewable
and supply. The simulator is available for the interested reader Energy Laboratory, 2006.
at www.power.uwaterloo.ca. [28] M. Eastment and R. Hendron, “Electricity and water consumption for
laundry washing by washing machine worldwide,” J. Energy Efficiency,
Springer Science vol. 3, pp. 365-382, 2010.
R EFERENCES [29] H. Hassen, “Implementation of energy hub management system for
residential sector,” Master’s thesis, University of Waterloo, April 2010.
[1] Ontario Power Authority, “Supply Mix Advice Report,” vol. 1, part 1-1,
Ontario, Canada, Dec. 2005. [Online]. Available:
http://www.powerauthority.on.ca/Storage/18/1338− Part− 1-
1− Supply− Mix− Summary.pdf
[2] Canada. “Comprehensive Energy Use Database,”
2008, 12 Sep 2011. [Online]. Available:
http://oee.nrcan.gc.ca/corporate/statistics/neud/dpa/trends− res− on.cfm
[3] D. and R. International, Ltd., “Buildings Energy Data Book 2011,” U.S.
Department of Energy, March 2012.