Content uploaded by Zunaira Nadeem
Author content
All content in this area was uploaded by Zunaira Nadeem on Jun 16, 2021
Content may be subject to copyright.
An intelligent opportunistic scheduling of home
appliances for demand side management
Zunaira Nadeem1, Nadeem Javaid2,∗, Asad Waqar Malik1, Abdul Basit Khan1,
Muhammad Kamran1, and Rida Hafeez1
1School of Electrical Engineering and Computer Science (SEECS), National
University of Sciences and Technology (NUST), Islamabad 44000, Pakistan;
2COMSATS Institute of Information Technology, Islamabad 44000, Pakistan
∗Correspondence: nadeemjavaidqau@gmail.com, www.njavaid.com
Abstract. Demand side management plays a vital role in load shifting
to off peak hours from on peak hours in response to dynamic pricing.
In this paper, we propose an optimal stopping rule (OSR) and firefly
algorithm (FA) for the demand response based on cost minimization.
Each appliance gets the best opportunistic time to start its operation
in response to dynamic electricity pricing. The threshold based cost is
computed for each appliance where each appliance has its own priority
and duty cycle regardless of their energy consumption profile. Numerical
simulations show that our proposed scheme performed well in lowering
cost, waiting time and peak to average ratio.
1 Introduction
The exponential rise in energy demand accompanied by the continuous decline
in energy generation requires an ongoing up gradation in todays energy infras-
tructure. Academia and research industry have considered it to be the serious
concern in order to address the future energy needs. The idea of a smart elec-
tricity system has moved the concept of the conventional grid to the smart grid
(SG). Smart grid impersonates a vision of the future electric generation sys-
tem integrated with the advanced sensing technology, two way communication
at transmission and distribution level in order to efficiently supply smart elec-
tricity in a smart way. Reliability, cost saving, self-healing, self-optimization,
consumer friendliness pollutant reduction are a few of the many benefits of SG.
This modern grid is motivated by several social, economic and environmental
factors.
Demand side management (DSM) is the modification of consumer energy
demand for energy in response to variation in electricity prices. DSM programs
consist of planning, implementing and monitoring activities of utility, particu-
larly designed to encourage consumers to modify their level of using electricity. In
literature, there are different DSM strategies are proposed such as load shifting,
valley filling, demand response (DR) etc. Demand response is defined as changes
in end-user electricity usage behaviour in response to the variation in electricity
2 Zunaira Nadeem et al.
prices over time. The DR program includes price based and incentive-based DR
programs.
In this study, we have proposed a mathematical technique optimal stopping
rule (OSR) compared with the meta-heuristic algorithm firefly (FA) which op-
portunistically schedule home appliances based on priority constraint. The op-
portunistic scheduling refers to the best starting time of an appliance based on
their priority. Each appliance has its own priority and length of operation time.
The appliance with high priority has high cost and lesser waiting time and vice
versa. So we can clearly see the trade off between electricity cost and waiting
time.
The simulation is conducted on single home appliances. Appliances are cate-
gorized into three categories: Shiftable appliances, Non Shiftable appliances and
uninterrupted appliances. The main objective of this work is the reduction in
electricity cost. In addition, the yearly cost saving of appliances are demon-
strated.
The remaining paper is organized in the following way. Section II describes
the related work. Section III explains the system model. Section IV presents the
simulation results. In the end, the whole summary is concluded in Section V.
2 Literature Review
Previously, the load scheduling problem was solved by non-integer linear pro-
gramming (NILP), linear programming (LP), integer linear programming (ILP)
and mixed integer linear programming (MILP). For example, in [1], a MILP is
formulated to solve the scheduling problem. MILP is precise and efficient, but
on the other hand, it is not capable of handling complex scenarios. The proposed
technique is implemented in the realistic scenario which helps price makers in
the realistic market. Today the world is more dependent on the SG so energy
management is one of the serious concerns of today’s world. In [2], the schedul-
ing problem is formulated for the household task which helps the consumers to
save their money. The system relies on different pricing schemes, local energy
generators and flexible task based on different deadlines.
In [3] the two of the heuristic techniques, shuffled frog leaping (SFL) and
teaching and learning based optimization (TLBO) algorithms are performed for
modeling energy consumption scheduling problem. The results show that the
proposed techniques performed well for cost reduction.The authors in [4] pro-
posed real-time opportunistic scheduling technique based on OSR. The schedul-
ing problem for home appliances is performed in centralized and distributed
pattern. The comparison of OSR and LP is performed and the simulation re-
sults depict that OSR performed well and it is less complicated in comparison
to LP. In the paper [5], the real time pricing (RTP) based environment is pre-
sented to solve opportunistic scheduling for home appliances using OSR. The
purpose of this work is user comfort while considering cost reduction. The mod-
ified first come first serve scheduling algorithm is proposed which compares with
An intelligent opportunistic scheduling of home appliances for DSM 3
first come first serve and early deadline first scheduling algorithms. The simula-
tions validate the performance of proposed techniques of cost and user comfort.
In [6], the energy management controller (EMC) is proposed based on heuristic
techniques: binary particle swarm optimization (BPSO), genetic algorithm (GA)
and ant colony optimization (ACO). Furthermore, multiple knapsack problem
is used to formulate the problem. The price schemes used for simulation are the
inclined block rate and time of use. In this paper, the users are categorized into
active and passive users. The desired objectives of the work are cost reduction,
peak to average ratio (PAR) minimization and user comfort maximization. The
simulations conducted show that GA based management controller worked well
than the ACO and BPSO based EMC. The authors in [7] consider three major
divisions; commercial, residential and industrial. A large set of several types of
controllable devices is handled to solve an optimization problem using heuris-
tic techniques. Results demonstrate that reasonable cost saving can be done by
considering PAR. In [8], the problem is highlighted to manage load without pay-
ing extra money, previously to manage load threshold limit is applied for each
home if the consumer cross that limits additional charges are applied in their
bill. To avoid this problem there is a need to propose an efficient load balancing
technique which can manage the load with minimum cost while reducing the
waiting time of electrical appliances. In this paper, a multi-objective optimiza-
tion technique is used for load balancing. The result reveals that the proposed
techniques efficiently minimize the electricity bill and reduce the waiting time of
appliances.
Zunaira et al. in [9] proposed the three hybrid techniques OSR-GA, OSR-
TLBO and OSR-FA. The simulations are conducted on single home as well as
multiple homes. The proposed techniques are formulated via chance constrained
optimization. The desired objective of this work is cost minimization while con-
sidering user comfort and PAR
3 System Model
The proposed system model consists of home area network (HAN), advanced
metering infrastructure (AMI), energy management controller (EMC), smart
meter (SM), RTP signals, smart appliances (SA) and programmable logic con-
troller (PLC). The clear picture of all these devices can be seen in Fig. 1.
3.1 HAN
A HAN is a network deployed with in the home that connects a person devices,
such as computers, telephone, video games, SM, home security system and all
those appliances that requires Wifi. HAN support wired and wireless technology
such as Zigbee, Wifi, WiMax, etc. Typically, HAN consists of a broadband inter-
net connection that connects multiple devices via a third party wired or wireless
modem. The EMC receives the pricing information from the power grid through
SM via an internet [2].
4 Zunaira Nadeem et al.
2 way communication
Communication way
Information Flow
Cloth Dryer Dishwasher
EMC
HAN
Smart Home
WIFI
Usage Data
RTP Signal
Smart Meter
Power Station
Refrigerator
Fig. 1. Proposed System Model
2 4 6 8 10 12 14 16 18 20 22 24
24 hours
0
0.02
0.04
0.06
0.08
0.1
0.12
Normalized hourly energy usage
Clothes dryer
Dishwasher
Refrigerator
Fig. 2. Energy Profile of Appliances
3.2 AMI
AMI allows the utility to get energy consumption information from the smart
home by a SM to utility. It is the two way communication between consumer
and utility that integrates SM, communication networks, and EMC.
An intelligent opportunistic scheduling of home appliances for DSM 5
Electricity cost is calculated according to the price signal provided by the
regulatory authority. In this work, the RTP price signal is used, which is more
realistic and give accurate results for load shifting to off peak hours from on
peak hours. In RTP, prices of every next hour vary while it remains the same
throughout the specific time slot.
3.3 EMC
An EMC is a digital device used to monitor the appliances and their energy
consumption rating. It contains all the information about appliances, power
rating, priorities, threshold, length of operation and based upon these parameters
and the inducted algorithm HEMC take a decision and provide low cost schedule
pattern for the appliances. In this paper, OSR and FA are considered as an
optimization algorithm.
3.4 SA
We divide a day into time slots, denoted as T and consider that there is a set
of different appliances, denoted as A1, A2, A3,...., An. Each appliance has its
own energy profile. Energy profile of each appliance can be shown in a Fig. 1.
The energy cycle of the appliance is for 24-hour time slots. Each time slot is
considered as 1 hour which is denoted by the parameter τ. In this paper, we are
concerned with Shiftable appliances. The three main appliances are cloth dryer,
dishwasher, and refrigerator.
The appliances which take more time to accomplish their task can take more
than one time slot. The dishwasher completes its task in three stages such as
main wash, final rinse, and heated dry. The cloth dryer has less than an hour
duty cycle. It accomplishes its task in only one cycle. The refrigerator has two
main stages ice-making and defrost, and only defrost stage of the refrigerator
can be scheduled for electricity cost reduction.
3.5 PLC
PLC helps to implement the optimization algorithm provided to it and it pro-
vides the visualization through the graphical interface between appliances, SM,
and the controller.
3.6 Problem Formulation
The objective function of the propose solution is where Tstands for electricity
cost:
minimize(
24
X
t=1
(
n
X
ap=1
T+W aitingT ime)) (1)
6 Zunaira Nadeem et al.
Cost The given equation is used to calculate the cost where Prhour is power
and Pap
ris electricity price and χ(τ) is on off status of appliances:
T=
T
X
hour=1
(
n
X
ap=1
(P rhour ×Pap
r×χ(τ))) (2)
Load Load can be computed by the equation below where tstands for load
and χ(τ) indicates the on off status of appliances:
t=
T
X
τ=1
n
X
ap=1
Pap
r×χ(τ) (3)
PAR PAR highlights the load peaks and helps to balance the load. It can be
formulated by equation below:
P AR =max(ετ
n)
1
TPT
τ=1 PAn
n(ετ
n)
∀T= 24 (4)
Threshold P is the electricity price uniformly distributed between [po,pp]
where ρois the lower limit and ρpis the maximum limit. The appliances are
turned on once their electricity price is lesser than the threshold. This equation
is taken from [4].
Z∗=r2(ρp−ρo)µτ
ε+ρo(5)
Waiting Time Waiting time or delay of the appliance is considered as the avg.
waiting time of an appliance. It can be computed by given equation where t´s is
the starting time of an appliance and tr is the requested time for an appliance:
W aitingT ime =tr −t
0
s(6)
4 Simulation Results and Discussion
We assume three shiftable appliances dishwasher, clothdryer and refrigerator for
scheduling. The initial parameters are given in Table 1 and the energy profile
of all appliances is shown in Fig. 2. All these appliances are plotted on two
different priorities. To avoid randomness we have taken 10 iterations of the whole
population.
4.1 Clothdryer
The Fig. 3 and Fig. 4 show the performance of cloth dryer in terms of average
cost and waiting time, respectively. The Fig. 5 shows the PAR of cloth dryer.
Cloth dryer has only one main cycle. The simulations for cloth dryer is summed
up in Table 2.
An intelligent opportunistic scheduling of home appliances for DSM 7
Table 1. Parameters of Appliances
Appliances Average Power
(kW)
LOT (hours) Priority (µ)
Clothdryer 3.0 0.75 [0.001,0.13]
Dishwasher 0.8 1.75 [0.001,0.015]
Refrigerator 0.089 24 [0.0033,0.0099]
Table 2. Cloths Dryer Simulation Statistics
Technique Priority Average
Cost/Month
Reduction in
Cost (%)
Average Delay in
Hours/Day
OSR 0.001
0.13
69.5299
151.9062
70
35
6.5706
3.5833
FA 0.001
0.13
129.7632
134.7632
45
43
8.6357
6.6758
1 2 3 4 5 6 7 8 9 10 11 12
Time (Month)
0
5
10
15
20
25
30
Cost ($)
Unscheduled
OSR µ=0.001
OSR µ=0.13
FA µ=0.001
FA µ=0.13
Fig. 3. Cost of Cloth Dryer
In Table 2, OSR and FA are compared with priority 0.001 and 0.13. In the
case of OSR, the cost is 35% reduced with priority i.e., 0.13 and 70% reduced
with priority i.e., 0.001. When we compare FA with unscheduled cost about 45%
of the cost is saved with priority 0.001 and 43% of the cost is saved with priority
1.13. If we compare both techniques, in this case, OSR performs better than
the FA. However, the delay time of OSR and FA for high priority i.e., 0.13 is
comparatively less. It clearly shows the trade off between waiting time and cost.
4.2 Dishwasher
The Fig. 6 and Fig. 7 illustrate the average cost and average waiting time of
dishwasher for each month. The result of the simulation is summarized in Table
3. We compare two techniques OSR and FA for dishwasher schedule. Our target
8 Zunaira Nadeem et al.
1 Cases
0
1
2
3
4
5
6
7
8
9
10
Waiting Time (Hours)
OSR µ=0.001
OSR µ=0.13
FA µ=0.001
FA µ=0.13
Fig. 4. Waiting Time of Cloth Dryer
Unscheduled OSR OSR FA FA
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
PAR
Unschedule
OSR µ=0.001
OSR µ=0.13
FA µ=0.001
FA µ=0.13
Fig. 5. PAR of Cloth Dryer
is to find the one which gives the best time schedule for a dishwasher with
minimum cost and delay. Simulation results demonstrate that 15.9%, 46% of
cost saving with a minimum delay of 2.02 and 4.3 respectively is calculated with
priority 0.015 and 62.49%, 47.71% of cost reduction with a minimum delay of
5.0 and 6.2 respectively is calculated when priority is 0.001. By comparing their
results, we noticed OSR perform optimal than FA in terms of average cost and
delay.The Fig. 8 shows the PAR of the dishwasher.
4.3 Refrigerator
The refrigerator is operated 24h a day. There are two states of refrigerator: Icing
state and defrost state we schedule the defrost stages of the Refrigerator. The
An intelligent opportunistic scheduling of home appliances for DSM 9
1 2 3 4 5 6 7 8 9 10 11 12
Time (Month)
5
10
15
20
25
30
Cost ($)
Unscheduled
OSR µ=0.001
OSR µ=0.015
FA µ=0.001
FA µ=0.015
Fig. 6. Cost of Dish Washer
1 Cases
0
1
2
3
4
5
6
Waiting Time (Hours)
OSR µ=0.001
OSR µ=0.015
FA µ=0.001
FA µ=0.015
Fig. 7. Waiting Time of Dish Washer
Table 3. Dishwasher Simulation Statistics
Technique Priority Average
Cost/Month
Reduction in
Cost (%)
Average Delay in
Hours/Day
0SR 0.001
0.015
88.1046
197.5074
62.49
15.9
5.0233
2.0244
FA 0.001
0.015
122.8069
124.6949
47.71
46
6.2783
4.3631
Fig. 9 and Fig. 10 show the average cost and average waiting time of refrigerator
for each month and Fig. 11 shows the PAR. In order to obtain the required
objectives, we use two techniques to see the effects on cost and average waiting
10 Zunaira Nadeem et al.
Unscheduled OSR OSR FA FA
0
0.5
1
1.5
2
2.5
3
3.5
PAR
Unschedule
OSR µ=0.001
OSR µ=0.015
FA µ=0.001
FA µ=0.015
Fig. 8. PAR of Dish Washer
1 2 3 4 5 6 7 8 9 10 11 12
Time (Month)
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
Cost ($)
Unscheduled
OSR µ=0.0033
OSR µ=0.0099
FA µ=0.0033
FA µ=0.0099
Fig. 9. Cost of Refrigerator
Table 4. Refrigerator Simulation Statistics
Technique Priority Average
Cost/Month
Reduction in
Cost (%)
Average Delay in
Hours/Day
OSR 0.0033
0.0099
51.2714
54.0010
17
15
6.1875
4.1875
FA 0.0033
0.0099
52
54
14
12
9.0625
7.0052
time. The summary of refrigerator is summed up in Table 4. Simulation results
demonstrate that TLBO with priority 0.0099 reduced 15% of cost with an aver-
age delay of 4.18 hours but in the other case when the priority of refrigerator is
An intelligent opportunistic scheduling of home appliances for DSM 11
1 Cases
0
1
2
3
4
5
6
7
8
9
10
Waiting Time (Hours)
OSR µ=0.0033
OSR µ=0.0099
FA µ=0.0033
FA µ=0.0099
Fig. 10. Waiting Time of Refrigerator
Unscheduled OSR OSR FA FA
0
0.5
1
1.5
2
2.5
3
PAR
Unschedule
OSR µ=0.0033
OSR µ=0.0099
FA µ=0.0033
FA µ=0.0099
Fig. 11. PAR of Refrigerator
0.0033, 17% of the cost is reduced with an average delay of 6.1 hours. Similarly,
FA is compared with unscheduled cost with priority 0.0099 it saves 12% cost and
when it decreases to priority 0.0033, it saves 14%. In the case of a refrigerator,
OSR performs well as it saves more cost than FA.
5 Conclusion
The heuristic techniques for residential DR based on the RTP scheme are pro-
posed in our paper. Three different types of home appliances with different pri-
orities and duty cycle are considered to solve the cost minimization problem
while considering the waiting time. OSR and FA are the proposed algorithms
12 Zunaira Nadeem et al.
and their performances are compared in terms of cost and waiting time. The
simulation results show that OSR performed better than the FA. Our proposed
techniques facilitate the residential customer to participate in DR program.
References
1. de la Torre, S., Arroyo, J.M., Conejo, A.J. and Contreras, J., 2002. Price maker self-
scheduling in a pool-based electricity market: a mixed-integer LP approach. IEEE
Transactions on Power Systems, 17(4), pp.1037-1042.
2. Derin, O. and Ferrante, A., 2010, April. Scheduling energy consumption with local
renewable micro-generation and dynamic electricity prices. In First Workshop on
Green and Smart Embedded System Technology: Infrastructures, Methods, and
Tools.
3. Derakhshan, G., Shayanfar, H.A. and Kazemi, A., 2016. The optimization of demand
response programs in smart grids. Energy Policy, 94, pp.295-306.
4. Yi, P., Dong, X., Iwayemi, A., Zhou, C. and Li, S., 2013. Real-time opportunistic
scheduling for residential demand response. IEEE Transactions on smart grid, 4(1),
pp.227-234.
5. Rasheed, M.B., Javaid, N., Ahmad, A., Awais, M., Khan, Z.A., Qasim, U. and Alra-
jeh, N., 2016. Priority and delay constrained demand side management in realtime
price environment with renewable energy source. International Journal of Energy
Research, 40(14), pp.2002-2021.
6. Rahim, S., Javaid, N., Ahmad, A., Khan, S.A., Khan, Z.A., Alrajeh, N. and Qasim,
U., 2016. Exploiting heuristic algorithms to efficiently utilize energy management
controllers with renewable energy sources. Energy and Buildings, 129, pp.452-470.
7. Logenthiran, T., Srinivasan, D. and Shun, T.Z., 2012. Demand side management
in smart grid using heuristic optimization. IEEE transactions on smart grid, 3(3),
pp.1244-1252.
8. Muralitharan, K., Sakthivel, R. and Shi, Y., 2016. Multiobjective optimization tech-
nique for demand side management with load balancing approach in smart grid.
Neurocomputing, 177, pp.110-119.
9. Nadeem, Z., Javaid, N., Malik, A.W. and Iqbal, S., 2018. Scheduling Appliances with
GA, TLBO, FA, OSR and Their Hybrids Using Chance Constrained Optimization
for Smart Homes. Energies, 11(4), p.888.
