A Novel Pricing Mechanism for Demand Side Load Management in Smart Grid

Abstract

This paper proposes a novel demand response (DR) mechanism based on real time electricity prices with the objective of cost reduction. The novelty of the proposed mechanism lies in the concept that electricity sub-prices are calculated based on the fraction of energy consumed by each unit/home. While, in traditional residential energy management programs, one DR signal applies to each associated unit without considering low, medium or high energy consumers. The proposed mechanism calculates the electricity prices for each individual unit based on which the electricity is calculated. Furthermore, the proposed mechanism is designed in such a way that it is equally feasible for all types of DR programs being used for demand side energy management. To assess the feasibility and practical applicability of the proposed mechanism, extensive simulations are conducted. The simulation results verify that the mechanism is efficient in calculating sub-prices without violating the utility constraint (i.e., the net cost remains same in both traditional and proposed mechanisms).

Full text

A novel pricing mechanism for demand
side load management in smart grid
Muhammad Babar Rasheed1,2, Nadeem Javaid1, Muhammad Awais3,
Mariam Akbar1, Zahoor Ali Khan4,∗
1COMSATS Institute of Information Technology, Islamabad 44000, Pakistan
2Laboratory for Intelligent Integrated Networks of Engineering Systems (LIINES),
Thayer School of Engineering, Dartmouth, Hanover, NH, 03755, USA
3Department of Technology, The University of Lahore, Lahore 54000, Pakistan
4Computer Information Science, Higher Colleges of Technology, Fujairah 4114, UAE
∗Correspondence: Zahoor Ali Khan; Email: zahoor.khan@hct.ac.ae
Abstract—This paper proposes a novel demand response (DR)
mechanism based on real time electricity prices with the objective
of cost reduction. The novelty of the proposed mechanism lies
in the concept that electricity sub-prices are calculated based
on the fraction of energy consumed by each unit/home. While,
in traditional residential energy management programs, one DR
signal applies to each associated unit without considering low,
medium or high energy consumers. The proposed mechanism
calculates the electricity prices for each individual unit based on
which the electricity is calculated. Furthermore, the proposed
mechanism is designed in such a way that it is equally feasible
for all types of DR programs being used for demand side energy
management. To assess the feasibility and practical applicability
of the proposed mechanism, extensive simulations are conducted.
The simulation results verify that the mechanism is efficient
in calculating sub-prices without violating the utility constraint
(i.e., the net cost remains same in both traditional and proposed
mechanisms).
Index Terms—smart grid; demand side management; home
energy management; renewable energy source; energy storage
system; real time pricing; genetic algorithm; binary particle
swarm optimization; wind driven optimization.
I. INTRODUCTION AND BACKGRO UND
Smart grid (SG) is envisioned as a novel electric power sys-
tem which includes advanced metering infrastructure (AMI)
having the capability of sensing the power consumption of end
users by integrating two way communication technologies [1]-
[3].
DR is one of the most important components/programs of
SG used for providing solutions for enhancing the efficiency
and reliability of future power systems [4]. Generally, DR
programs are designed where end users reschedule their energy
consumption patterns during high peak hours in response to
time varying electricity prices. Here the objective is two-fold:
(i) minimization of end users electricity cost, and (ii) grid
stability [1], [5]. However, these DR programs can only be
implemented by encouraging customer by proving incentives
in the form of bill reduction or lower electricity unit price.
In this regards, much work has been done by different au-
thors to effectively design the DR programs with the objective
of cost reduction and grid stability. In [6], a mixed integer
linear programming (MILP) technique is used to solve the
problem of DR program [7]. However, most of existing studies
use day ahead pricing which is known in advance to residential
customers. Although, this scheme is widely implemented and
efficiently used, however, it is not giving more optimal results.
This is because the hourly real time price (RTP) fluctuates
and provides different unit prices in comparison to day ahead
prices [8]. Recently, Stackelberg game based techniques are
popular for solving hierarchical decision problems in power
systems [9]. In [10], [11], authors use Stackelberg based game
theoretical approach to model and solve the trading problem
between users and service provider with the aim at maximizing
the utility benefits through optimal design of day ahead DR.
Later on, this DR program is used to induce end users to
reschedule their power consumption to minimize net bill. Prior
to this, game theory based optimization techniques are also
designed in two different perspectives: (i) game between users
and utilities, and (ii) game among residential users, only. In
former one, game theory based techniques are developed to
help utilities to design optimal DR programs which later on
are useful in reducing end user electricity cost. The later
one focusses on the development of techniques inside user
premisses where the electricity price is given as a input.
Users schedule their load according to the given input without
taking into consideration utility constraints [12], [13]. Here,
the existing models are developed on the basis of day ahead
prices, once the optimal strategy is found, each player have to
follow that strategy.
In [14], [15], authors use optimal stoping rule (OSR) theory
and RTP to schedule residential load. Each load has been
assigned a time factor based on which a threshold is calculated.
Then based on this threshold, it is decided that weather the
load is ON or OFF in next time slot. Note that, electricity price
is considered real time which is uniformly distributed over the
given limits. A RTP based Stackelberg game theoretical tech-
nique has been proposed in [16]. This scheme is better than
aforementioned Stackelberg game based techniques because
it uses RTP instead of day ahead. A one-leader, N-follower
game is designed and formulated where each player helps
2017 31st International Conference on Advanced Information Networking and Applications Workshops
978-1-5090-6231-7/17 $31.00 © 2017 IEEE
DOI 10.1109/WAINA.2017.119
283

in adopting the optimal strategy. Through simulations, it has
been shown that this technique comparatively achieves most
optimum results. Until now, all the optimization techniques
discussed above focus in reducing electricity cost and grid
stability. However, no technique considers the consumption
levels of end users which is still a challenging area. So, the
current work considers the highlighted problems are solves
this problem to benefit both, consumers and utilities.
The remaining paper is organized into the following sec-
tions. Section II gives the motivation behind the proposed
work. Section III provides the details about various DR
programs and their importance. The impact of DR programs
in residential energy management perspectives is discussed
in section IV. Problem formulation of both traditional and
proposed methods is given in section V. At the end, simulation
results and conclusion are given in sections VI and VII,
respectively.
II. MOTIVATION
The SG concept has come to incorporate different com-
munication protocols, distributed energy resources, AMI, and
DR programs to efficiently manage the electricity load. In
this concept, residents are provided with the facility to cut
their electricity bill by rescheduling the load patterns. In this
regards, various energy management and scheduling schemes
based on DR programs have been found in the literature
[1]- [11]. Among these schemes, the major focus is towards
electricity cost reduction in customer premisses and peak to
average ratio reduction in utility premisses. As discussed in
literature, residential energy management programs usually
depend on many factors such as customer preferences, weather
conditions and DR programs. Among these factors, DR is one
of the key factors which is the base of energy management pro-
grams. However, in traditional energy management programs
[1]- [13], day ahead pricing (DAP), time of use (TOU) pricing,
critical peak pricing (CPP), or day ahead RTP schemes are
used for any particular region/area. Normally, the electricity
unit price in these DR programs is calculated on the basis
of aggregated energy consumption of all associated loads as
shown in Fig. 1. Whereas, it is very difficult to operate more
than one electricity retailers in the same region. However, the
problem associated with these DR programs is that the end
users having minimum energy consumption are penalized. Be-
cause, DR programs are designed for a particular region rather
than any particular home. More importantly, this problem can
be more severe if more number of users consume less amount
of energy during any given time spam. In this situation, high
energy consumers and electricity retailers are not penalized.
Because, retailers receive electricity bills based on consumed
energy and high energy consumers pay normal bills according
to DR. Although, they have to pay more electricity bill due
to more consumption. So, this is the underlying problem for
both end users and utilities. Up-to our knowledge which is
based on extensive literature review, this problem has not been
highlighted yet. Considering this problem and its eventual
drawbacks on low consumption users, this work proposes a
new mechanism which calculates electricity bills of end users
according to their usage of electricity. For example, if any
user consumes more energy in comparison to other users, it
will get new pricing for next particular hour without affecting
the aggregate revenue of energy retailers. The novel aspect
of this work is that every time when any user gets new price
based on energy consumption, the aggregate bill remains same.
The details of the proposed DR program is discussed in the
following sections.
III. CHARACTERIZING DR
Defined broadly, the term DR refers to the participation of
energy consumers in wholesale electricity market, analyzing
and responding to time varying prices. DR can also be defined
more precisely in relation to electricity perspective; changes
in electricity consumption patterns made by end users from
their normal patters in response to real time electricity prices
varying over fixed intervals of time, or to provide monetary
incentives majorally designed to induce lower consumption
during high electricity prices or the conditions when the
reliability of power system jeopardized [17].
A. Importance of DR
Due to physical constraints and limitations, electricity can-
not be stored at large scale power systems. In this regards,
the total production of electricity must be equal or more
than net demand at that time spam. In addition, the marginal
cost of electricity is extremely variable due to continuous
fluctuations in energy demand of end users. While, the cost of
electricity varies over some fixed intervals of time (e.g., every
15 min., 1 hour), where most customers face retail electricity
prices which represent average electricity production including
transmission and distribution costs. So, the disconnection
between short term marginal electricity cost and retail cost
paid by consumers leads inefficient use of available resources.
This is because, consumers cannot visualize the short term
cost of supplying electricity and eventually they have no or
little incentives to participate in DR programs. Thus, the flat
electricity prices encourage the consumers to over consume
electricity during hours when electricity prices are higher than
the average prices and under consume when prices are lower
relative to average rate. In result, the electricity prices may
be higher because some times, electric generators must turn
ON to fulfill non responsive energy demand of end users.
This will allow the power producers to raise electricity prices
above average market prices in certain conditions. So, in the
long term, the inefficient DR can result in expensive energy
generation. And, the efficient DR programs can reduce the
risks of capacity investment in peaking generation plants.
B. Types of DR
Currently, there are two available DR options: (i) price
based DR programs, and (ii) incentive based DR programs.
The detailed discussion of these DR programs is given below.
284

TABLE I
COMPARISON OF VARIOUS DR PROGRAMS OFFERED BY POLICYMAKERS/RETAILERS
Price based DR Incentive based DR
TOU: a rate having variable electricity prices with
different slots of time (usually 24 hours having
equal time slot of one hour). TOU pricing reflects
the electricity unit price of the delivering power
to the end users. TOU rates vary by time of a
day (e.g., high peak, mid peak and low peak).
TOU tariffs are widely used by industrial and
residential consumers.
RTP: a rate where electricity price fluctuates
hourly depending on the changes in energy whole
sale market. Generally, these prices are known in
advance to customers on day ahead or hourly basis.
CPP: are designed in such a way that high
unit price is predefined over certain intervals of
time. CPP can be triggered due to high prices or
by high system contingencies by utilities obtaining
power from wholesale markets. CPP is not itself
a pricing scheme. It can be superimposed to any
other pricing scheme like RTP, TOU, DAP, etc.
Usually, it is implemented for a limited time
duration due to high energy demand or long term
system faults. Although, CPP scheme is not very
popular. However, participating customers get
incentives during non CPP hours.
Direct load control (DLC): is a program in which energy retailers or
providers directly shut down some residential loads on short notices in
order to address reliability contingencies. Utilities are given a full access
to partial loads (e.g., air-conditioners and water heaters ). Participating
customers obtain incentives from retailers. In case of violating the
agreement of participation, customers are imposed penalties. DLC
programs are used for small industrial and residential customers.
Interruptible/curtailable service (I/C): these programs are usually
integrated to ongoing customer tariffs and provide incentives or discounts
in participating the load reduction during fault conditions. Customers that
not not participate this program during fault conditions are charged high
bills or remove from the system for that particular time period. Normally,
these programs are offered to large industries or commercial sectors.
Demand Bidding: in these programs: (i) customers offer the utilities the
prices at which they agree to curtail their load, (ii) encourage the end
users how much load they would curtail on the given utility price.
Emergency DR: programs that offer incentives to the customers
who are willing to participate in load reductions during fault conditions.
Penalties for violating customers may or may not be used.
Capacity Market Programs: are designed for those customers who
are willing to curtail pre-specified load during given time duration.
1) Price based DR Programs: The programs include TOU,
flat rate pricing (FRP), DAP, CPP, and RTP which can also be
characterized as price based DR programs. In these programs,
the cost of electricity unit fluctuates within certain limits
in according to the variation in energy demand. Because,
power producers have to maintain balance between demand
and supply. Where, high energy demand increases the tariff
rate and vice versa. Furthermore, DR programs are offered
as an alternative to fixed rates. The end users can take
benefits of these programs by properly responding by adjusting
their consumption schedules. Usually, customer responses are
driven by an internal decision making process where any kind
of load modification are permitted.
2) Incentive based DR Programs: Incentive based DR
programs refer contractual arrangements especially designed
by energy producers, retailers, policy makers, grid operators,
etc., to elicit energy demand from customers during certain
hours. According to these programs, participating customers
are given monetary incentives which are other than the regular
fixed or average prices. More importantly, the participation in
these programs are open for every customer. But, the cus-
tomers violating the programs after participating, levy penal-
ties. In order to measure the capacity of demand reduction, DR
programs typically use a method to establish baseline energy
level. Table I provides the details and comparison of all DR
programs.
IV. ANALYZING DR IN ENERGY MANAGEMENT
PERSPECTIVE
Regarding home energy management, various DR based
energy management techniques have been proposed in the
literature [1]- [13]. In these techniques, the major focus
is towards residential load/appliance scheduling using price
based DR programs. Usually, optimization based intelligent
algorithms are designed which embed in home energy man-
agement controller to perform the assigned tasks. Based on
some input parameters, these algorithms work autonomously
without human involvement. These inputs include:
•electricity price (DR signal) which is directly obtained
via smart metering infrastructure from utility or energy
retailer.
•customer preferences based on which the designed algo-
rithm performs intelligent decisions.
•environmental parameters such as temperature, human
presence, illuminance intensity, etc. These input values
can be obtained from sensor and actuator networks.
Among aforementioned inputs, electricity price signal is one
of the most important factors in residential energy management
domain. While, in most of the energy management techniques,
TOU or day ahead RTP pricing schemes are mostly used.
This is because these schemes are widely adopted by various
energy retailers and hence easily implementable in residential
sector. As discussed in sections II and III, DR programs are
designed for a particular region rather than for each home
285

which is impractical in nature. In this way, electricity unit
price is calculated on aggregated energy consumption basis
which is same for that particular region. Although, this method
for calculating energy unit price seems practical and being
implemented in almost everywhere. However, low energy
consumption users get penalized as they are getting the same
prices (aggregate based) even consuming less energy. As we
stated earlier, it is difficult to design a separate price signal
for each home and is impractical in nature. Hence, in light
of these trade-offs, this work focusses on the issue that how
can we minimize this problem. Where, low energy consumers
and high energy consumers can pay the bill according to their
actual consumption without affecting the utility.
V. P ROBLEM FORMULATION
In this section, we first discuss the electricity cost calcula-
tion using traditional method based on RTP. Then we discuss
how the proposed mechanism is used to calculate sub-prices
using the same electricity prices obtained form utility.
A. Traditional Method
In this section, we formulate load optimization problem
using input parameters. Lets there are unumber of con-
sumers/residential units having variable energy demand Eand
comfort requirements.
Ut=[u1,u
2,u
3, ....un],(1)
Where, Utis the set of all residential units. The energy
demand of these units is given as below:
•Eu1=u1
•Eu2=u2
•Eu3=u3
•Eun=un
where, Eu1denotes the energy demand of u1unit and vice
versa. The total energy demand of unusers in time slot tis
given as:
Euu=n
u=1 Eu≈Eq,∀t. (2)
Where, Eunis total energy consumption which is assumed
to be equal to Eqquantity. The electricity cost of a single user
can be calculated using the following equation:
cu1=n
u=1 Eu1×ept,(3)
Where, eptdenotes the electricity unit price in time t. The
total electricity cost of unusers based on energy consumption
Eunis given as:
Cun=n
u=1 Eun×ept≈Cq,∀[t∈T],(4)
where, Cqis the total cost due to aggregate energy consump-
tion. Here, one important point is that eptis fixed during time
duration tin case of TOU or day ahead RTP. The drawback
associated with this scheme is that energy prices set by utilities
are not based upon individual energy consumer. But on the
basis of aggregate energy consumption for any particular
region (section III). Due to this strategy, the low, medium
and high energy consumers are charged equally which is not
feasible for low and medium energy consumption users. To
mitigate or overcome this problem, there must be a mechanism
which charged electricity prices to the associated consumers
on the basis of their energy consumption. Fig. 1 gives the
overview of the traditional energy management method where
electricity prices are provided to each unit, directly.
B. Proposed Method
Considering the aforementioned problem, we have proposed
a novel pricing mechanism which charges electricity prices
based upon the fraction of energy consumption. Eq. 3 gives
the expression for electricity cost based on fixed interval
electricity price. However, according to the proposed method,
the energy consumption cost can be calculated using the
following equations:
ept1=Eu1×k,
ept2=Eu2×k,
.
.
.
eptn=Eun×k,
(5)
where, kis a constant which decides the electricity unit price
for each users based on their associated energy consumption.
The value of kis calculated using the following expression:
k=n
u=1 Eu×ept
n
u=1(Eu)2.(6)
Now, the new electricity unit price neptu1of u1can be
calculated as:
neptu1=k.(Eu1)2.(7)
The total cost of all users unis calculated as follows:
neptun=n
u=1 k.(Eu)2.(8)
Note: according to fig. 2, we add an aggregator in retailer
block. Because, we cannot design new electricity price once
obtained form retailer or energy distributor. Based on this price
and energy consumption data received from smart meter, we
calculate the unit price accordingly (eq. 8). However, it might
also be possible that we can add an extra device/controller to
perform the same work which we proposed.
VI. SIMULATION RESULTS
Before the discussion of simulation results, table II gives the
parameters used in the proposed work. Here, the electricity
prices and appliance date are obtained from [18], [19]. For
simulation studies, we consider four different cases for the
validation of our proposed technique.
case-1: in this case, we consider 9.83 $/KWh electricity
price and perform simulations. Fig. 3a shows the comparison
of Poand Pn. It is clear form figure that Pois fixed for each
286

Fig. 1. Residential energy management mechanism using traditional pricing scheme, (SM-smart meter; EMC-energy management controller)
Fig. 2. Residential energy management mechanism using proposed pricing scheme
TABLE II
APPLIANCE POWER RATINGS,ELECTRICITY PRICES AND AGGREGATE COST
Electricity price ($) 9.83 8.63 8.87 12.00 – –
Load (KWh) 2.5 3.00 2.00 2.5 3.5 1.5
Aggregate cost
(traditional) ($)
147.45 233.01 133.05 180.00 – –
Aggregate cost (proposed)
($)
147.45 233.01 133.05 180.00 – –
home as used in traditional schemes. While, Pnis designed in
such a way that its price is different for each home which is
based on the individual energy consumption. In fig. 4b, energy
consumption cost of each home based on Poand Pnis shown.
In home-1, Pois a little bit more as compared to Pn.Soas
the electricity cost is also comparatively less. Home-5 has high
power rating 3.5 KWh and based on Po, its total cost is less as
compared to Pn. Because, Pnis designed in such a way the
unit price is calculated on the basis of net energy consumption
of that particular house/unit. So, the relative unit price is high
287

123456
Homes
0
2
4
6
8
10
12
14
Cost ($)
P
o
P
n
(a) Comparison of old (Po) and new (Pn), ept
123456
Homes
0
5
10
15
20
25
30
35
40
45
50
Cost ($)
P
o
C
o
P
n
C
n
(b) Cunbased on Poand Pn
Fig. 3. Analyzing the impact of Poand Pnon the Cun(case-1: ept= 9.83 $/KWh), (power rating of each home = h1−2.5,h
2−3,h
3−2,h
4−2.5,h
5−
3.5,h
6−1.5)
123456
Homes
0
2
4
6
8
10
12
14
16
Cost ($)
P
o
P
n
(a) Comparison of Poand Pn,ept
123456
Homes
0
10
20
30
40
50
60
70
80
Cost ($)
P
o
C
o
P
n
C
n
(b) Cunbased on Poand Pn
Fig. 4. Analyzing the impact of Poand Pnon the Cun(case-2: ept= 8.63 $/KWh), (power rating of each home = h1−2.5,h
2−2.5,h
3−4.5,h
4−
4.5,h
5−6.5,h
6−6.5)
for this home. However, the aggregated cost of electricity is
same as calculated on the basis of price obtained form utility
which is given in table II.
In case-2: the eptand power requirement of each home is
different. In this case, the energy demand of first two homes
is same and the energy demand of next two homes are also
same and vice versa. Fig. 4a shows the Poand Pnand it is
clear that unit price for each pair of homes having same energy
demand is same. Similarly, fig. 4b shows the comparison of
energy consumption cost based on Poand Pn, respectively.
From this figure, you can analyze that new unit price for each
home depends on the fraction of energy consumed by each
home. Low energy consumption home receives low unit price
and high energy consumption home receives the high unit
price. The remaining case-3 and case-4 give the validation
of the proposed mechanism using different electricity prices
and power ratings (figs. 5, 6). In each case, we consider
different consumption levels and unit prices in order to prove
the effectiveness of the proposed idea.
Remarks: From the simulation results, it is clear that using
the proposed mechanism, we can reduce extra bills paid by res-
idential consumers due to aggregated DR programs designed
by utilities/retailers. It is also proved that net electricity bill
based on utility based DR signal and the total electricity bill
calculated using the proposed schemes is always same. This
is the evidence that the proposed technique can be used as a
benchmark for residential DR without violating any constraint.
288

123456
Homes
0
2
4
6
8
10
12
14
Cost ($)
P
o
P
n
(a) Comparison of Poand Pn,ept
123456
Homes
0
5
10
15
20
25
30
35
40
45
50
Cost ($)
P
o
C
o
P
n
C
n
(b) Cunbased on Poand Pn
Fig. 5. Analyzing the impact of Poand Pnon the Cun(case-3: ept= 8.87 $/KWh), (power rating of each home = h1−1.5,h
2−3,h
3−4,h
4−0.5,h
5−
3.5,h
6−2.5)
123456
Homes
0
2
4
6
8
10
12
14
16
Cost ($)
P
o
P
n
(a) Comparison of Poand Pn,ept
123456
Homes
0
10
20
30
40
50
60
70
Cost ($)
P
o
C
o
P
n
C
n
(b) Cunbased on Poand Pn
Fig. 6. Analyzing the impact of Poand Pnon the Cun(case-4: ept= 12.00 $/KWh), power (rating of each home = h1−1.5,h
2−4,h
3−4,h
4−
0.5,h
5−2.5,h
6−2.5)
VII. CONCLUSION
In this paper, we have proposed a novel DR mechanism
for residential load management with the aim of energy cost
reduction. For this purpose, optimization problem is formu-
lated mathematically and expressions to calculate electricity
sub-prices are developed. Then based on RTP, both traditional
and proposed methods are used for electricity cost reduction.
Simulation results show that the proposed mechanism reduces
the same electricity cost as reduced by traditional method.
However, the major difference is that the proposed method
calculates different electricity prices for each unit based on its
total energy consumption.
In future, we have plans to implement the proposed cost cal-
culation mechanism with different demand side management
algorithms to provide the benefits to the end users. Moreover,
we are also interested in implementing this mechanism in real
time environments.
REFERENCES
[1] M. Yu and S. H. Hong, “A Real-Time Demand-Response Algorithm
for Smart Grids: A Stackelberg Game Approach,” in IEEE Transac-
tions on Smart Grid, vol. 7, no. 2, pp. 879-888, March 2016. doi:
10.1109/TSG.2015.2413813.
[2] D. G. Holmberg, OpenADR Advances, ASHRAE Journal. [On-
line]. Available: http://www.escholarship.org/uc/item/3dn3426x, ac-
cessed Oct. 17, 2016.
[3] A. Zakariazadeh, S. Jadid, and P. Siano, “Stochastic operational
scheduling of smart distribution system considering wind generation
and demand response programs,” Int. J. Elect. Power Energy Syst.,
vol. 63, pp. 218225, Dec. 2014.
289

[4] Z. Quanyan, P. Sauer, and T. Basar, “Value of demand response in the
smart grid,” in Proc. 2013 IEEE Power Energy Conf. Illinois (PECI),
Champaign, IL, USA, pp. 7682.
[5] A. Zakariazadeh, S. Jadid, and P. Siano, “Smart microgrid energy and
reserve scheduling with demand response using stochastic optimiza-
tion,” Int. J. Elect. Power Energy Syst., vol. 63, pp. 523533, Dec.
2014.
[6] A. Agnetis, G. de Pascale, P. Detti, and A. Vicino, “Load scheduling
for household energy consumption optimization,” IEEE Trans. Smart
Grid, vol. 4, no. 4, pp. 23642373, Dec. 2013.
[7] O. Erdinc, N. G. Paterakis, T. D. P. Mendes, A. G. Bakirtzis and J.
P. S. Catalao, “Smart Household Operation Considering Bi-Directional
EV and ESS Utilization by Real-Time Pricing-Based DR,” in IEEE
Transactions on Smart Grid, vol. 6, no. 3, pp. 1281-1291, May 2015,
doi: 10.1109/TSG.2014.2352650.
[8] S. Althaher, P. Mancarella and J. Mutale, “Automated Demand Re-
sponse From Home Energy Management System Under Dynamic
Pricing and Power and Comfort Constraints,” in IEEE Transactions
on Smart Grid, vol. 6, no. 4, pp. 1874-1883, July 2015, doi:
10.1109/TSG.2014.2388357.
[9] W. Tushar et al., “Three-party energy management with distributed
energy resources in smart grid,” IEEE Trans. Ind. Electron., vol. 62,
no. 4, pp. 24872498, Apr. 2015.
[10] F.-L. Meng and X.-J. Zeng, “A Stackelberg game-theoretic approach
to optimal real-time pricing for the smart grid,” Soft Comput., vol. 17,
no. 12, pp. 23652380, 2013.
[11] S. Maharjan, Z. Quanyan, Z. Yan, S. Gjessing, and T. Basar, “Depend-
able demand response management in the smart grid: A Stackelberg
game approach,” IEEE Trans. Smart Grid, vol. 4, no. 1, pp. 120132,
Mar. 2013.
[12] Z. M. Fadlullah, D. M. Quan, N. Kato, and I. Stojmenovic, “GTES: An
optimized game-theoretic demand-side management scheme for smart
grid,” IEEE Syst. J., vol. 8, no. 2, pp. 588597, Jun. 2014.
[13] N. Yaagoubi and H. T. Mouftah, “User-Aware Game Theo-
retic Approach for Demand Management,” in IEEE Transactions
on Smart Grid, vol. 6, no. 2, pp. 716-725, March 2015, doi:
10.1109/TSG.2014.2363098.
[14] M. B. Rasheed, N. Javaid, A. Ahmad, M. Awais, Z. A. Khan, U.
Qasim, and N. Alrajeh, “Priority and delay constrained demand side
management in real-time price environment with renewable energy
source,” Int. J. Energy Res., 40: 20022021, 2016, doi: 10.1002/er.3588.
[15] Y. Peizhong, D. Xihua, I. Abiodun, Z. Chi, L. Shufang, “Real-time
Oppertunistic Scheduling for Residential Demand Response”, IEEE
Transactions on Smart Grid, vol. 4, no. 1, pp. 227-234, 2013.
[16] M. Yu and S. H. Hong, “A Real-Time Demand-Response Algorithm
for Smart Grids: A Stackelberg Game Approach,” in IEEE Transac-
tions on Smart Grid, vol. 7, no. 2, pp. 879-888, March 2016, doi:
10.1109/TSG.2015.2413813.
[17] QDR, Q. “Benefits of demand response in electricity markets and
recommendations for achieving them.” US Dept. Energy, Washington,
DC, USA, Tech. Rep (2006).
[18] Yi, Peizhong, X. Dong, A. Iwayemi, C. Zhou, and S. Li, “Real-
time opportunistic scheduling for residential demand response”, IEEE
Transactions on Smart Grid 4, no. 1: 227-234, 2013. 2, pp. 716-725,
March 2015, doi: 10.1109/TSG.2014.2363098.
[19] http://www.wholesalesolar.com/solar-information/how-to-save-
energy/power-table, accessed Oct. 17, 2016.
290