ChapterPDF Available

A New Memory Updation Heuristic Scheme for Energy Management System in Smart Grid

Authors:

Abstract and Figures

In the last decade, high energy demand is observed due to increase in population. Due to high demand of energy, numerous challenges in the existing power systems are raised i.e., robustness, stability and sustainability. This work is focused for the residential sector Energy Management System (EMS), especially for the smart homes. An EMS is proposed which shifts the electricity load from high price to low price hours. To fulfill the high load demand of electricity consumers, we have proposed a new Memory Updation Heuristic Scheme (MUHS), which efficiently schedule the appliance from on peak to off peak hours. The objective of our new scheme MUHS is to automate the EMS. The significance of our new proposed MUHS scheme shown the efficiency by reducing Cost, Peak to Average Ratio (PAR) and increase User Comfort (UC) by balancing the load demand in peak times.
Content may be subject to copyright.
A New Memory Updation Heuristic
Scheme for Energy Management System
in Smart Grid
Waleed Ahmad1, Nadeem Javaid1(B
), Sajjad Khan1, Maria Zuraiz2,
Tayyab Awan3, Muhammad Amir1, and Raza Abid Abbasi1
1COMSATS University, Islamabad 44000, Pakistan
nadeemjavaidqau@gmail.com
2COMSATS University Islamabad, Abbotabad Campus, Abbotabad, Pakistan
3National University of Science and Technology, Islamabad, Pakistan
http://www.njavaid.com
Abstract. In the last decade, high energy demand is observed due to
increase in population. Due to high demand of energy, numerous chal-
lenges in the existing power systems are raised i.e., robustness, stability
and sustainability. This work is focused for the residential sector Energy
Management System (EMS), especially for the smart homes. An EMS
is proposed which shifts the electricity load from high price to low price
hours. To fulfill the high load demand of electricity consumers, we have
proposed a new Memory Updation Heuristic Scheme (MUHS), which
efficiently schedule the appliance from on peak to off peak hours. The
objective of our new scheme MUHS is to automate the EMS. The sig-
nificance of our new proposed MUHS scheme shown the efficiency by
reducing Cost, Peak to Average Ratio (PAR) and increase User Comfort
(UC) by balancing the load demand in peak times.
Keywords: Smart homes ·Demand side management ·
Energy Management System ·Appliance scheduling ·
Critical Peak Pricing
1 Introduction
In the last couple of years, high energy consumption in residential areas [1], has
been observed due to an increase in household electric appliances. The energy
demand has increased upto 27% due to the increase in high power consumption
appliances, for example: appliances used for cooling and heating systems. These
appliances are the main reason of energy shortage. The traditional utilities have
insufficient resources to fulfill the today’s load demand. Therefore, more power
generation is required to provide comfort to the users. The Home Energy Man-
agement System (HEMS) is adopted along with Demand Response (DR) to the
cope with peak load by scheduling the high-power electric appliances by shifting
c
Springer Nature Switzerland AG 2020
L. Barolli et al. (Eds.): AINA 2019, AISC 926, pp. 53–66, 2020.
https://doi.org/10.1007/978-3-030-15032-7_5
54 W. Ahmad et al.
the On-peak hours power load to the Off-peak hours. In this regards, the inter-
mittent energy resources brought into consideration in [2]. Demand Side Energy
Management System (DSEM) in [3] is proposed with Direct Load Controlling
(DLC) integrated with Dynamic Pricing (DP) model.
The power supply companies started to manage the load centrally by schedul-
ing the smart appliances to reduce electricity bill and Peak to Average Ratio
(PAR) in congested time intervals. The conventional grid system is unstable
due to lack of control and unidirectional communications. To achieve these chal-
lenges and remove the flaws of the conventional system, Smart Grid (SG) in
[4] has emerged which is comprised of bidirectional communication, fault toler-
ance, and real time self support recovery. SG is facilitated with Advance Metering
Infrastructure (AMI) that helps in bidirectional communication between the end
users and service provider. SG has two important key components: DSM and DR.
With the help of DSM, the utility companies are providing consumers’ aware-
ness programs to use the electricity wisely. In DR schemes, the utilities motivate
the consumers to actively engage with the SG to minimize the load demand and
electricity bill in the pre-defined budget. The consumers have reduced the energy
cost through DR schemes efficiently [5,6].
2 Related Work
In the last few years, the researchers have made a tremendous work in SG to
decrease the energy cost for the consumers using different optimization tech-
niques. To balance the load and reduce the energy cost with the increasing
demand are fascinating and tough research problems, that have been addressed
by the experts along with their efforts for solving it. Several DSM methods have
been shown previously to minimize the users electricity bill and waiting time,
while maximizing individuals comfort. In [7], an Integer Linear based Program-
ming (ILP) scheme was proposed, which is used to control the energy supply in
the residential area. This proposed technique successfully achieved the desired
goal by shifting the high power appliances load to the OFF-peak hours. In [8],
Genetic Algorithm (GA) was proposed. The aim of the proposed algorithm was
to minimize the price and load demand equilibrium; both in the demand and
supply side. In [9], the authors presented heuristic techniques, Particle Swarm
Optimization (PSO) with GA. This work shows the reduction in peak power
demand and decreased the energy cost. This technique is employed three price
tariff plans: Time Of Use (ToU), Real Time Pricing (RTP) and Critical Peak
Pricing (CPP) by using Multi Knapsack Problem (MKP). Their proposed opti-
mization scheme efficiently achieved the goal by lesson the energy cost and PAR.
Simulations also revealed that GA is more efficient than PSO. In [10], Multiple
Integer Linear Programming (MILP) is used with HEMS to tackle the load
demand using Renewable Energy Resources (RES), i.e., electric mobiles, and
energy storage systems. In [11], the authors present a model for cost reduction
in residential sector by using Mixed Integer Non-Linear Programming (MINLP)
by including ToU pricing model. This proposed method minimizes the cost of
A New Memory Updation Heuristic Scheme for EMS in Smart Grid 55
Table 1. Related work schemes comparison
Technique(s) Achievement(s) Limitation(s)
ILP The proposed scheme achieved cost
reduction
UC is neglected
PSO GA and
MKP
This model reduces cost and PAR UC is ignored
GA Simulation results show decrements
of cost and PAR
UC is ignored and simulation
performed on residential area
only
DP Minimization of cost Authors did not mention
equipments cost and
maintenance
MILP and
heuristic
algorithm
Proposed scheme successfully
manages the load equilibrium
Neglected the cost factor
MILP This technique efficiently
minimized the cost
Par while UC is ignored
GA-DSM Model shows reduction in
electricity consumption
Authors did not mention
PAR and UC
MINLP The proposed scheme accomplished
desired results by reducing the cost
PAR is not considered in
proposed work
GA and BPSO This approach shows minimization
in the cost and PAR
UC is neglected and focused
only on smart home
residential areas
CSA The proposed scheme efficiently
schedules the shift able appliances
while considering users load
demand
PAR and UC did not
presented in their research
work
energy by the total of 25%. However, PAR is not evaluated in their research
work. GA along with Binary PSO algorithm is proposed in [12].
This optimization scheme reduce the user’s electricity bill and PAR. Simu-
lations show effectiveness of the described model. In [13], the authors propose
knapsack bounds scheme to lower the peak load and minimize the rebound peaks
of the different ON-peak time intervals. In [14], HEMS is used with DR which
offers strategies to optimize the home appliances operations. The scheduling of
appliances is a challenging task and needs the optimal solutions. HEMS focuses
on UC as well as cost minimization; however, these two factors have trade off,
i.e., in order to reduce cost, User Comfort (UC) has to be sacrificed. In lit-
erature, it is often observed that HEMS works as a single objective or multi-
objective models. In this paper, the single objective model is used to decrease
the cost. This problem is solved by Gradient Base Repair Optimization (GBRO)
technique. A new model, an Evolutionary Accretive Comfort Algorithm (EACA)
56 W. Ahmad et al.
[13], has been introduced which depends on postulations that permits the time
and device based priorities. Evaluation is based on input data which consist
of devices power rating, ToU, and absolute UC. The EACA has the capability
to develop the electricity usage pattern on pre-determined consumers budget.
The results from the experiment reveal that the EACA optimization technique
efficiently reduces the cost and increases UC. In [15], a Jaya-based algorithm
is proposed, which evaluates the power levels of distributed electricity sources,
minimize the actual power drop and reduces the production expenses. Simula-
tion results are compared with other optimization techniques like: Strawberry
Algorithm (SA) and Enhanced Differential Evolution (EDE).
3 Problem Statement
In residential sector, DSM has many challenges like devices scheduling, irregular
usage of energy, consumers negligence for the defined schedules and infeasible
solutions for increasing the demands. To stabilize the consumers demands, many
DSM techniques are brought into consideration from the last few years. The
irregular nature of residential side demand causes many problems on supply
side management. Heuristic schemes are considered as one of the cost-effective
solutions. More generation of electricity is required which in turn raises the
production cost. For this purpose, an optimized system is required to schedule
smart devices in the residential side to schedule from ON-peak hours to OFF-
peak hours with less waiting time. In [16], the purpose of appliance scheduling is:
load balancing, and cost and PAR reduction. The above discussion clearly reveals
that price and PAR can be reduced by applying some optimization techniques
as presented in [17]. In this scheme, an EMS is designed for the efficient energy
management scheduling in peaks. Energy optimization of the heating and cooling
systems is also considered in one of the most important objectives of this work
because they consume a large portion of energy (for example, 64% of the total
residential energy in Canada).
4SystemModel
Many researchers struggled on DSM and Supply Side Management (SSM). The
main focus of our research is on DSM in a domestic area. Every smart home
in the domestic area has Smart electrical appliances and Smart Meters (SMs)
which are connected to the EMC. SMs performs bidirectional communication
between Smart homes and Utility. EMC receive all the information from smart
devices and sensors inside a smart house by using a home area network. EMC
evaluates the data received from smart devices and calculates the total average
of the load. This required load demand after evaluation transfer to the utility
through SMs. After that, the utility satisfies the consumer demand and transfer
pricing scheme of electricity via SMs to the consumer. SMs then assign all these
pricing information and energy to EMC. So, EMC has now load demand and
power rating of smart devices and pricing tariffs collected from the utility. Now,
A New Memory Updation Heuristic Scheme for EMS in Smart Grid 57
Fig. 1. Proposed system model
the smart appliances are scheduled by EMC, based on load demand and pricing
schemes received from smart homes and utility. EMC performs the scheduling
function based on algorithms installed in it. In our proposed model, we have
created a new scheme MUHS based EMC. Alongside we also created Grey Wolf
Optimization (GWO) based EMC and Crow Search Algorithm (CSA) based
EMC. We have compared the efficiency of all three schemes and shown that
our proposed Scheme successfully achieved the desired goal i.e., minimize Peak
Average Ration (PAR) and Maximize user comfort. All the bidirectional traffic
between EMC to SMs and SMs to the utility is performed by Wi-Fi routers via
wide area network. Devices with their similar type, energy rating and Length of
Operational Time (LOT) are given in Table 2.
In our model, we have deemed 15 smart appliances from a smart home. With
the opinion associated with electricity payment, we used CPP value signal per-
taining to payment calculation. In this paper, we break 1-day time into 24-h
time intervals. The actual cause of this specific action could be that, the waiting
time of some smart appliances are usually equal to 1 h schedule. We have parti-
tioned smart home appliances into 24-h schedule. A lot of the home appliances
deemed within this paper are usually working continually for an hour. Due to
this, we avoid the actual wastage associated with time and acquire 24-h time
schedule. In all earlier mentioned information, the actual electricity repayment
and also delay time from the smart appliances are usually decreased. With that,
our model is more cost effective and robust. Our goal of this activity is to lower
the power consumption throughout peak time slots to diminish total power cost
and to reduce waiting time (Fig. 1).
58 W. Ahmad et al.
Now, mathematically formulation of afore named goals, we have mathemat-
ical equations here. Total power utilization is expressed applying Eq. 1.
Load =
24
t=1
pS(t),S(t)=[1/0] (1)
PAR is defined in Eq. 2and total expense for 24-h is figured in Eq. 3.
PAR =(Max(Load)/Avg (Load)) (2)
Cost =
24
t=1
()EP pS(t)),S(t)=[1/0] (3)
The Eq. 1shows power rating of devices, and S(t) represent the ON and OFF
status of the devices, where 1 indicates that appliance is ON at a particular
time interval and 0 indicates a device is OFF. In Eq. 2, PAR is measured as the
maximum load from 24-h power utilization form and then split it by the average
of the same 24 h energy utilization pattern. The EP in Eq. 3shows the energy
price for a 24-h time slot.
During this paper, the actual smart appliances presumed are usually divided
throughout a few distinct different types like (Delay-Tolerant, Delay-Intolerant
and Delay-Intolerant with Flexible Load.
4.1 Appliances Classification
4.1.1 Delay-Tolerant Appliances
Appliances that can be interrupt any time in 24-h time slot are delay-tolerant
shown in Table 1. Shifting the appliance time slot from On-peak hour to Off-peak
hour can decrease cost, PAR while UC may be sacrificed.
Table 2. Delay-tolerant appliances
Type Appliances
AC 1.5
Refrigerator 1.66
Iron 1
Vacuum cleaner 0.7
Water Heater 5
Dishwasher 1.32
Water Pump 1
A New Memory Updation Heuristic Scheme for EMS in Smart Grid 59
4.1.2 Delay-Intolerant Appliances
Delay-Intolerant is those electrical smart devices that cannot be suspended while
they are in running condition. These kind of appliances can be shift from one
time interval to another time interval before their working time.
4.1.3 Delay-Intolerant Base Appliances
Base appliances are those electrical devices which cannot be stopped while they
start working. These kind of appliances depends upon consumer demand.
4.2 Price Tariff
Several types of pricing schemes have been implemented around the globe,
Through the help of these pricing tariffs, the consumer can schedule their appli-
ances from On-Peak to Off-peak time intervals. In our proposed scheme, We
have implemented CPP pricing scheme and achieved our objective by reducing
PAR, and maximize the UC by reducing in their delay time. The main focus is
to balance the energy consumption during peak price times to decrease the cost.
However, cost and waiting time always have a trade-off between them (Table 3).
Table 3. Appliances used in simulations
Appliances Power rating (kwh) LOT (h)
Refrigerator 1.666 24
Vacuum Cleaner 0.7 0.7
Waterpump 1 8
Washing Machine 1.4 3
Cloth dryer 5 4
Dishwasher 1.32 3
Water Heater 5 8
Iron 2.4 3
AC 1.5 8
Cooker 0.225 4
Toaster 0.8 1
Printer 0.011 2
Light 0.18 12
Blender 0.3 2
Oven 2.4 4
60 W. Ahmad et al.
5 Meta-heuristic Algorithms
Main-stream optimization methods are mostly numerical algorithms like, LP,
Convex Programming (CP) and ILP, etc. are not doing properly for large quan-
tity of devices due to their slow computation. Therefore, we go for meta-heuristic
methods for wise home appliances scheduling to accomplish our major aim that
will be energy cost minimization and minimize delay time. We choose meta-
heuristic rather than heuristic and mathematical practices since these are rela-
tively efficient than mathematical practices while working with a large number of
devices and these are the problem independent, whereas heuristic practices are
issue dependent, this means heuristic practices are made to resolve a particular
problem. The selected meta-heuristic calculations are mentioned in more detail
in the next sections.
5.1 CSA
A new meta-heuristic optimizer in CSA [18] is proposed by Alireza Askarzadeh
in 2016. It is inspired by the smart behavior of storing excess food by crows.
CSA is a meta-heuristic population centered algorithm which operates on the
basis of the indisputable fact that crows collect their additional food in hiding
locations for later use. Taking a food stock which is concealed by a crow is
a difficult task since if crow note that another crow is following it, the crow
fools another crow by setting its path and planning to a different location as
opposed to the actual food resource concealed by it. From relating them with
the optimization method, the crows would be the searchers, the surroundings
could be the research space, the location of the surroundings can relate with the
possible answer, fitness function is evaluated based on food source quality. In
whole search space the best food origin is considered as Best solution.
5.2 GWO
GWO [19], Meta-Heuristic algorithm is a new hunting technique of grey wolves
inspired by the leadership chain. There are four types of wolves incorporated in
method used by the leadership chain of grey wolves. Furthermore we find there
are three main hunting phases which are, searching for the prey, attacking over
the prey and encircling the prey. In compliance with the social hierarchy of grey
wolves, fittest solution to be practiced is the alpha, afterwards the beta, delta
and omega comes accordingly.
5.3 MUHS
In this section, we define our proposed Meta-heuristic algorithm MUHS. In the
MUHS, we have used initialization parameters and generate random population.
Now, in MUHS, population updation is completely based on fittest solutions. On
the basis of these best solutions, we have updated and modify the search agents
A New Memory Updation Heuristic Scheme for EMS in Smart Grid 61
Algorithm 1. Algorithm for MUHS
1: Start
2: Initialize all parameters Max iter: (NP,D,A)
3: Randomly generate initial population Xi(i=1,2, ..., n)
4: X(i, j)=rand(NP,D);
5: ittr =0, transformation probability
6: while ittr < M aximum I teration do
7: for i=24toPopulation do
8: Let Yis a random position of search space
9: Evaluate the position of the A
10: Initialize the memory of each A
11: if x(i, :) l&x(i, :) uthenx(i, :) = x(i, :)
12: transf orm local candidate solution ;
13: else
14: transf orm global candidate solution ;
15: end if
16: end For
17: Identif y the best and worst candidate solutions
18: if Solution moves towards the best then
19: U pdate solution ;
20: else
21: M odify sol ution v ia mutation ;
22: end if
23: if New solution moves towards worst solutions then
24: U pdate solution ;
25: else
26: Discard new solution and keep old solution ;
27: end if
28: end while
according to environments. The storage or population operation is performed by
utilizing the comparison of the fitness of every single option of the population
with the fitness of storage positions, and then, when the condition holds true,
the storage of the population is modified by the best option which is originated
in the last result of main positions in MUHS. Therefore, we choose the updation
with its position, review and update memory by finding the best fittest criterion.
Our goal is to find more optimized solution by having large number of random
data to achieve best results. Following are the 5 steps.
1. Initialization of the population and parameters
2. Probability checking of the new environment
3. Encompassing the solution
4. Review and update memory
5. Finding the best fitness criteria.
62 W. Ahmad et al.
Fig. 2. CPP pricing scheme
Fig. 3. Per hour load
Fig. 4. PAR
A New Memory Updation Heuristic Scheme for EMS in Smart Grid 63
Fig. 5. Hour cost
6 Simulation and Reasoning
As this part of our discussion depict the results of our simulation in detail.
By incorporating the simulations, algorithms are ratify on the bases of PAR,
delay time, energy consumption and energy cost. We have incorporated 15 sin-
gle home appliances, for the simulation. For calculation of electricity bills we
have used CPP pricing method. We have shown CPP pricing signal in Fig. 2.
For each unscheduled slot Fig.3shows the consumption pattern as well as sched-
uled scenarios. The proposed MUHS technique results indicates that scheduling
of power in better way. The overall energy consumption of MUHS is less than
other techniques to avoid the peak creation at any slot of the day. It is proposed
that MUHS optimization technique is better in shifting the load from peak time
intervals to off peak interval time. User satisfaction is affected by the load shift-
ing while it gives benefits to the user as in cost reduction. DSM just not only
limits its usefulness to consumers but also beneficial for utility as well. Utility is
assisted by the decrease in PAR to keep is stability which leads to cost reduc-
tion. The performance of different algorithms optimization in the form of PAR
is shown in Fig. 4which clearly shows the proposed MUHS outperformed the
other techniques in terms of PAR. In Fig. 5, CSA, GWO and MUHS per hour
cost is shown. Plots indicates the peak hour reduction in scheduling scheme than
un-scheduling by shifting the load from peak price intervals to low price intervals.
Total cost production ratio of scheduled and unscheduled is presented in
Fig. 6, which clearly indicate the efficiency of our optimized algorithms in total
cost. However CSA and GWO beats our proposed scheme in term of cost because
our main goal was to achieve PAR and waiting time. It is worth saying that our
proposed scheme successfully beats the unscheduled case very efficiently. The
consumer satisfaction and delay time is measured in waiting time shown in Fig. 7.
The UC is assumed as, the particular user waiting time in which user turn on the
appliances. Therefore we can say that UC is opposite to delay time. The plots in
figure clearly indicate that proposed scheme efficiently reduce the overall cost and
waiting time, so it is worth mentioning that our proposed scheme outperformed
other described schemes in term of waiting time. Figure 8represents the total
64 W. Ahmad et al.
Fig. 6. Cost
Fig. 7. Waiting time
Fig. 8. Total load
power load of scheduling and unscheduled appliances which clearly indicates the
total load is balance in all cases.
7 Conclusion
In our propose work, we have deployed DSM with EMC to manage the load
demand by shifting the smart devices from On-peak to Off-peak hours. We have
considered two heuristic schemes i.e., GWO and CSA. We proposed a new heuris-
tic scheme MUHA, which efficiently lower the PAR with considerable amount of
A New Memory Updation Heuristic Scheme for EMS in Smart Grid 65
cost and also increase UC as compared to GWO and CSA. Appliance classifica-
tion and power rating helps EMC to control and schedule the appliance according
to user satisfaction. However, from the simulation results, performance is evalu-
ated and shows that our new scheme MUHA outperformed than GWO and CSA
in term of PAR and delay-time.
References
1. Cena, G., Valenzano, A., Vitturi, S.: Hybrid wired/wireless networks for real-time
communications. IEEE Ind. Electron. Mag. 2(1), 8–20 (2008)
2. Ganji Tanha, M.: Security constrained unit commitment reserve determination in
joint energy and ancillary services auction (Doctoral dissertation)
3. Strbac, G.: Demand side management: benefits and challenges. Energy Policy
36(12), 4419–4426 (2008)
4. Gellings, C.W.: The concept of demand-side management for electric utilities. Proc.
IEEE 73(10), 1468–1470 (1985)
5. Chaabene, M., Ben Ammar, M., Elha jjaji, A.: Fuzzy approach for optimal energy
management of a domestic photovoltaic panel. Appl. Energy 84(10), 992–1001
(2007)
6. Pradhan, V., Balijepalli, V.M., Khaparde, S.A.: An effective model for demand
response management systems of residential electricity consumers. IEEE Syst. J.
10(2), 434–445 (2016)
7. Khan, M.A., Javaid, N., Mahmood, A., Khan, Z.A., Alrajeh, N.: A generic demand
side management model for smart grid. Int. J. Energy Res. 39(7), 954–964 (2015)
8. Samadi, P., Wong, V.W.S., Schober, R.: Load scheduling and power trading in
systems with high penetration of renewable energy resources. IEEE Trans. Smart
Grid 7(4), 1802–1812 (2016)
9. Agnetis, A., de Pascale, G., Detti, P., Vicino, A.: Load scheduling for household
energy consumption optimization. IEEE Trans. Smart Grid 4(4), 2364–2373 (2013)
10. Bradac, Z., Kaczmarczyk, V., Fiedler, P.: Optimal scheduling of domestic appli-
ances via MILP. Energies 8(1), 217–232 (2014)
11. Ullah, I., Javaid, N., Khan, Z.A., Qasim, U., Khan, Z.A., Mehmood, S.A.: An
incentive based optimal energy consumption scheduling algorithm for residential
users. Procedia Comput. Sci. 52, 851–857 (2015)
12. Yalcintas, M., Hagen, W.T., Kaya, A.: An analysis of load reduction and load
shifting techniques in commercial and industrial buildings under dynamic electric-
ity pricing schedules. Energy Build. 88, 15–24 (2015)
13. Khan, A., Javaid, N., Khan, M.I.: Time and device based priority induced comfort
management in smart home within the consumer budget limitation. Sustainable
cities and society (2018)
14. Khalid, A., et al.: Cuckoo search optimization technique for multi-objective home
energy management. In: International Conference on Innovative Mobile and Inter-
net Services in Ubiquitous Computing. Springer, Cham (2017)
15. Samuel, O., Javaid, N., Ashraf, M., Ishmanov, F., Afzal, M., Khan, Z.: Jaya-based
optimization method with high dispatchable distributed generation for residential
microgrid. Energies 11(6), 1513 (2018)
16. Khalid, A., Javaid, N., Guizani, M., Alhussein, M., Aurangzeb, K., Ilahi, M.:
Towards dynamic coordination among home appliances using multi-objective
energy optimization for demand side management in smart buildings. IEEE Access
6, 19509–19529 (2018)
66 W. Ahmad et al.
17. Marzband, M., et al.: Real time experimental implementation of optimum energy
management system in stand alone microgrid by using multi-layer and colony opti-
mization.Int.J.Electr.PowerEnergySyst.75, 265–274 (2016)
18. Mirjalili, S., Mirjalili, S.M., Lewis, A.: Grey wolf optimizer. Adv. Eng. Softw. 69,
46–61 (2014)
19. Askarzadeh, A.: A novel metaheuristic method for solving constrained engineering
optimization problems: crow search algorithm. Comput. Struct. 169, 1–12 (2016)
... Many scholars have conducted fruitful research into energy management of smart cities [17][18][19][20]. Hussain et al. [21] reviewed the status quo of the digital transform of China's oil and gas industry, summarized the problems in the development of the traditional energy industry, i.e., limited capability of research and development (R&D), uneven levels of informatization, information redundancy and resource waste, and emphasis on development over operation and maintenance, and suggested promoting the digital transform of modern energy industry chain through Internet of Things (IoT) Platform Plus and Energy Business Type Plus. ...
Article
Smart cities mark the shift from the traditional model of urban construction to the planned construction of a composite system between smartness and energy. Considering the defects of renewable energies, e.g., intermittency, randomness, and low dispatchability, it is imperative to explore the digital and unified management of urban renewable energies. Therefore, this paper presents a multi-objective optimization algorithm for digital management, which can quantify the multiple energy models of smart cities. Firstly, the dimensions of renewable energy construction in smart cities were detailed, and the functions, hierarchy, and data flows of the digital management system for renewable energies were plotted in turn. After that, the output probability models of typical renewable energy power generation systems were established, plus the objective functions of digital management for renewable energies. Finally, particle swarm optimization (PSO) was combined with genetic algorithm (GA) for multi-objective optimization of the digital management for renewable energies in smart cities. The proposed models and algorithm were proved effective through experiments.
... Many researchers working on the residential area to control and manage the household appliances. In [1], authors proposed a novel Memory Updation Heuristic Scheme (MUHS), which efficiently schedule the appliances from onpeak to off-peak time slots. We studied the literature work in price forecasting and feature selection. ...
Chapter
Full-text available
The vital part of the smart grid is electricity price forecasting because it makes grid cost saving. Although, existing systems for price forecasting may be challenging to manage with enormous price data in the grid. As repetition from the feature cannot be avoided and an integrated system is needed for regulating the plans in price. To handle this problem, a new price forecasting system is developed. This proposed model particularly integrated with three systems. Initially, features are selected from the random data by combining the Mutual Information (MI) and Random Forest (RF). The Grey Correlation Analysis (GCA) is used to remove the redundancy from the selected features. Secondly, the Recursive Feature Elimination (RFE) scheme is used to reduce the dimensions. Finally, classification is done based on Enhanced-Convolutional Neural Network (E-CNN) classifier to forecast the price. The simulation results show that our accuracy of the proposed system is higher than existing benchmark schemes.
... Electricity price forecasting is done by time series prediction and machine learning algorithms. In [35], authors proposed a model to foretell the day ahead Electricity price according to load demand and temperature, with the usage of neural network design and evaluation of the unique spectrum. The authors in [36], presented a probabilistic system to determine hourly price rate, where the bootstrapping framework is used for examining uncertainty and induced extreme learning machine strategy is introduced for wavelet neural networks. ...
Thesis
Full-text available
In the last decade, high energy demand is observed due to increase in population. Every home has now large number of smart appliances for the daily routine operations. Due to high demand of energy, numerous challenges in the existing power systems are raised, i.e., robustness, stability and sustainability. The objective of our new schemes Random Cell Elimination Scheme (RCES) and Memory Updation Heuristic Scheme (MUHS) is to automate the Energy Management System (EMS), so that the consumer gets facilitated. Price forecasting is performed to reduce the burden on utility and also ease the consumers. This work is focused on the residential sector EMS, especially for the smart homes. To fulfill the high load demand of electricity consumers, we have defined a fitness criterion. The researchers are working on new approaches to enhance and improve the power demand. The increasing demand of electricity creates peaks on utility. Therefore an improved Home Energy Management System (HEMS) is necessary for the automation of smart home to reduce the cost and peaks on utility. We have utilized three pricing systems, Time of Use (ToU) and Real Time Pricing (RTP) and Critical Peak Pricing (CPP) in our experiments. Furthermore, The vital part of the smart grid is electricity price forecasting because it makes grid cost effective. Although, existing systems for price forecasting may be challenging to manage with enormous price data in the grid. As repetition from the feature cannot be avoided and an integrated system is needed for regulating the plans in price. To handle this problem, a new price forecasting system is developed. This proposed model particularly integrated with three systems. Initially, features are selected from the random data by combining the Mutual Information (MI) and Random Forest (RF). The Grey Correlation Analysis (GCA) is used to remove the redundancy from the selected features. Secondly, the Recursive Feature Elimination (RFE) scheme is used to reduce the dimensions. Finally, classification is done based on Enhanced-Convolutional Neural Network (E-CNN) classifier to forecast the price. The experimental results shows that the proposed schemes performed better than mentioned benchmark schemes in reducing PAR and user discomfort and but also increase the price forecasting accuracy.
Article
Full-text available
This paper presents a model for optimal energy management under the time-of-use (ToU) and critical peak price (CPP) market in a microgrid. The microgrid consists of intermittent dispatchable distributed generators, energy storage systems, and multi-home load demands. The optimal energy management problem is a challenging task due to the inherent stochastic behavior of the renewable energy resources. In the past, medium-sized distributed energy resource generation was injected into the main grid with no feasible control mechanism to prevent the waste of power generated by a distributed energy resource which has no control mechanism, especially when the grid power limit is altered. Thus, a Jaya-based optimization method is proposed to shift dispatchable distributed generators within the ToU and CPP scheduling horizon. The proposed model coordinates the power supply of the microgrid components, and trades with the main grid to reduce its fuel costs, production costs, and also maximize the monetary profit from sales revenue. The proposed method is implemented on two microgrid operations: the standalone and grid-connected modes. The simulation results are compared with other optimization methods: enhanced differential evolution (EDE) and strawberry algorithm (SBA). Finally, simulation results show that the Jaya-based optimization method minimizes the fuel cost by up to 38.13%, production cost by up to 93.89%, and yields a monetary benefit of up to 72.78% from sales revenue.
Article
Full-text available
IIn this paper, we propose a home energy management system which employs load shifting strategy of demand side management to optimize the energy consumption patterns of a smart home. It aims to manage the load demand in an efficient way to minimize electricity cost and peak to average ratio while maintaining user comfort through coordination among home appliances. In order to meet the load demand of electricity consumers, we schedule the load in day-ahead and real-time basis. We propose a fitness criterion for proposed hybrid technique which helps in balancing the load during On-peak and Off-peak hours. Moreover, for real-time rescheduling, we present the concept of coordination among home appliances. This helps the scheduler to optimally decide the ON/OFF status of appliances in order to reduce the waiting time of appliance. For this purpose, we formulate our realtime rescheduling problem as knapsack problem and solve it through dynamic programming. This study also evaluates the behavior of the proposed technique for three pricing schemes including: time of use, real-time pricing and critical peak pricing. Simulation results illustrate the significance of the proposed optimization technique with 95% confidence interval.
Conference Paper
Increasing demand of power and emergence of smart grid has gain maximum attention of researchers which has further opened new opportunities for Home Energy Management System (HEMS). HEMS under Demand Response (DR) helps to reduce the On-peak hour load by shifting the load toward the Off-peak hours. This load shifting strategy effects the user comfort, however in return DR gives them incentives in term of electricity bill reduction. Consumer electricity cost and peak load have a tradeoff, to sort out this situation an efficient system is required. In this paper, we present a multi-objective HEMS to schedule home appliances using Cuckoo Search Algorithm (CSA) while considering the objective load fitness criteria. This proposed load fitness criteria effectively reduces the cost and peak load. Simulations are performed to verify the generic behavior i.e., system performance on any price tariffs. For this purpose, results are validated for three price signals: day-ahead Real Time Peak Price (RTP), Time of Use (TOU) and Critical Peak Price (CPP).
Conference Paper
Increasing demand of power and emergence of smart grid has gain maximum attention of researchers which has further opened new opportunities for Home Energy Management System (HEMS). HEMS under Demand Response (DR) helps to reduce the On-peak hour load by shifting the load toward the Off-peak hours. This load shifting strategy effects the user comfort, however in return DR gives them incentives in term of electricity bill reduction. Consumer electricity cost and peak load have a tradeoff, to sort out this situation an efficient system is required. In this paper , we present a multi-objective HEMS to schedule home appliances using Cuckoo Search Algorithm (CSA) while considering the objective load fitness criteria. This proposed load fitness criteria effectively reduces the cost and peak load. Simulations are performed to verify the generic behavior i.e., system performance on any price tariffs. For this purpose, results are validated for three price signals: day-ahead Real Time Peak Price (RTP), Time of Use (TOU) and Critical Peak Price (CPP).
Thesis
This study presents the method in which the energy and ancillary services auction are simultaneously cleared in the electricity market. By the security-constraint unit commitment model proposed in this study Independent System Operators (ISO) can determine the sufficient amount of reserve which is necessary to maintain security and Reliability of the system. Before the fixed reserve requirement either equal to a percentage of the system peak load or a thermal unit with the highest capacity is considered in energy and ancillary service auction in market clearing. The disadvantage of this method is high cost and insufficiency. When it is insufficient the system operator needs to committee more thermal units or does the load curtailment. At the time the fixed quantity is more than needed customers pay more although it is not necessary. Here the sufficient amount of reserve is determined based on the contingency which has been simulated. Contingencies include thermal unit outage and line outage is considered. The amount of reserves is obtained based on thermal units’ physical constraints and the rate offered in the electricity market. Then the integration of wind generation and its effects on the quantity of the reserve determination is considered. Since the wind power generation brings uncertainties to the power system we need to consider scenarios. In order to generate wind power generation scenarios, we use Monte Carlo simulation. Since the number of scenarios are too much and increase the problem complexity we use Fast backward/forward scenario reduction. This problem is solved through direct optimization problem including the minimization of the operational cost as well as satisfying the network security constraints when contingency happens.
Article
In this paper, an algorithm for energy management system (EMS) based on multi-layer ant colony optimization (EMS-MACO) is presented to find energy scheduling in Microgrid (MG). The aim of study is to figure out the optimum operation of micro-sources for decreasing the electricity production cost by hourly day-ahead and real time scheduling. The proposed algorithm is based on ant colony optimization (ACO) method and is able to analyze the technical and economic time dependent constraints. This algorithm attempts to meet the required load demand with minimum energy cost in a local energy market (LEM) structure. Performance of MACO is compared with modified conventional EMS (MCEMS) and particle swarm optimization (PSO) based EMS. Analysis of obtained results demonstrates that the system performance is improved also the energy cost is reduced about 20% and 5% by applying MACO in comparison with MCEMS and PSO, respectively. Furthermore, the plug and play capability in real time applications is investigated by using different scenarios and the system adequate performance is validated experimentally too.
Article
Introduction of real-time pricing models for the price of electricity may lead to a reduction in both economic and environmental burden compared with existing models. Thanks to the timely response to price changes during the day the user can achieve a reduction in the payments for energy. However, recent studies show that lack of knowledge of users and their unwillingness to adapt their habits to changing energy prices is the biggest obstacle to successful implementation of the model. In our work, we propose to solve this problem by introducing automatic optimal energy consumption scheduling framework. This article presents the authors extension in the past created MILP model for the optimization of household appliances. The model is extended by rules of behavior that allows preserving logical links in the course of periodically triggered optimization (so called receding horizon). We also present and test the software simulator, in which the designed model is implemented. The application simulates the operation of Building Energy Manager, which based on user preferences, prediction electricity prices and other important parameters controls the operation of appliances in the house.
Article
In this paper, we focus on the problems of load scheduling and power trading in systems with high penetration of renewable energy resources (RERs). We adopt approximate dynamic programming to schedule the operation of different types of appliances including must-run and controllable appliances. We assume that users can sell their excess power generation to other users or to the utility company. Since it is more profitable for users to trade energy with other users locally, users with excess generation compete with each other to sell their respective extra power to their neighbors. A game theoretic approach is adopted to model the interaction between users with excess generation. In our system model, each user aims to obtain a larger share of the market and to maximize its revenue by appropriately selecting its offered price and generation. In addition to yielding a higher revenue, consuming the excess generation locally reduces the reverse power flow, which impacts the stability of the system. Simulation results show that our proposed algorithm reduces the energy expenses of the users. The proposed algorithm also facilitates the utilization of RERs by encouraging users to consume excess generation locally rather than injecting it back into the power grid.