Modeling and optimization of Demand side Management in ......v List of Publications 1. Eity Sarker, Pobitra Halder, Mehdi Seyedmahmoudian, Elmira Jamei, Ben Horan, Saad Mekhilef, Alex

Modeling and Optimization of Demand Side Management in Smart Grid

by

Eity Sarker B.Sc. (Hons) in Telecommunication and Electronic Engineering

A thesis submitted in fulfilment of the requirements for the degree of Master of Engineering

School of Software and Electrical Engineering

Faculty of Science, Engineering and Technology

Swinburne University of Technology

Hawthorn, VIC 3122, Australia

July 2020

i

Declaration

I am affirming that the works presented in this thesis is my own work and have not been used

by any person for any other degree. To the best of my knowledge, this thesis does not contain

any materials which have been previously published, unless where due acknowledgement and

reference are provided.

Eity Sarker

July 2020

ii

Acknowledgements

I would like to thank my principal supervisor, Senior Lecturer Mehdi Seyedmahmoudian,

who helped me throughout my candidature. His continuous supports and encouragements

help me to complete this thesis.

I also acknowledge the support and motivation of my associate supervisor Professor Alex

Stojcevski. He helped me a lot to overcome the difficulties when I changed my principal

supervisor.

I am also grateful to my husband, Pobitra Halder for his continuous support and inspiration. I

also acknowledge the continuous support from my family.

At the end, I would like to thank my senior and junior fellows at Swinburne and Melbourne

who made my journey enjoyable during and outside of my study. I would like to

acknowledge the Swinburne University of Technology for providing me Tuition Fee

Scholarship and logistic supports for conducting the research.

iii

Dedicated to my husband and my parents

iv

Abstract

Demand side management is considered as a key technique that could address the issues of

increasing energy demand and environmental pollution as well as help to achieve

socioeconomic sustainability. It can also facilitate residents’ transfer into smart homes and

sustainable cities. However, demand side management is required to overcome a number of

challenges in terms of energy transmission, distribution, and effective utilization of energy

resources. In order to overcome these challenges, researchers are developing new model and

upgrading the existing smart grid model. Therefore, algorithms are used to solve the

optimization problems of these models. The aim of this thesis work is to minimize electricity

consumption from the main grid and maximize the use of renewable energy resources to

reduce the total energy cost of the consumer.

In this study, the loads were scheduled based on flexible pricing and time of use pricing tariff

rate employing binary particle swarm optimization (BPSO) algorithm in MATLAB platform.

The microgrid was mathematically modelled and analyzed for different households in terms

of electricity cost reduction. The total renewable energy from solar PV and wind turbine was

estimated based on the Victorian solar data and wind data.

The findings of the case study suggested that the BPSO based load scheduling strategy

provided more stable cost curve compared to that of genetic algorithm based scheduling

strategy. The analyses showed the potential benefits of implementation of demand response

in terms of reduction of peak load and electricity cost for all the case studies. As a result of

both demand response program and renewable energy integration, the consumers required a

minimal amount of electricity from the grid.

v

List of Publications

1. Eity Sarker, Pobitra Halder, Mehdi Seyedmahmoudian, Elmira Jamei, Ben Horan, Saad

Mekhilef, Alex Stojcevski (2020). Progress on the demand side management in smart grid

and optimization approaches. International Journal of Energy Research. John Wiley &

Sons. IF 3.34 (Q1)- Accepted. DOI: 10.1002/er.5631

2. Eity Sarker, Mehdi Seyedmahmoudian, Elmira Jamei, Ben Horan, Alex Stojcevski

(2020). Optimal management of home loads with renewable energy integration and

demand response strategy. Energy. Elsevier. IF 5.54 (Q1)- Under revision.

3. Eity Sarker, Mehdi Seyedmahmoudian, Ben Horan, Alex Stojcevski (2020). Optimum

scheduling of residential, industrial and commercial loads using BPSO algorithm.

(Extension of conference paper). Draft ready.

4. Eity Sarker, Mehdi Seyedmahmoudian, Ben Horan, Alex Stojcevski (2019). Optimal

scheduling of appliances in smart grid environment using BPSO algorithm. 11th

International Conference on Applied Energy. Västerås, Sweden.

vi

Table of Contents Declaration ........................................................................................................................... i

Acknowledgements .............................................................................................................. ii Abstract .............................................................................................................................. iv

List of Publications .............................................................................................................. v

List of Figures................................................................................................................... viii

List of Tables ...................................................................................................................... ix

Abbreviations ...................................................................................................................... x

CHAPTER 1: INTRODUCTION ....................................................................................... 1

1.1 Energy and sustainability ...................................................................................................... 1

1.2 Smart grid .............................................................................................................................. 2

1.3 Demand side management ..................................................................................................... 2

1.4 Algorithms for solving optimization problems...................................................................... 3

1.5 Problem statement and research questions ........................................................................... 4

1.6 Aims and objectives ............................................................................................................... 4

1.7 Outline of the thesis ............................................................................................................... 5

CHAPTER 2: LITERATURE REVIEW ........................................................................... 6

2.1 Introduction ........................................................................................................................... 6

2.2 Techniques and approaches of DSM ..................................................................................... 7

2.3 Challenges of DSM implementation .................................................................................... 11

2.4 Progress of DSM models and applications of algorithms ................................................... 15

2.5 Integration of renewable energy sources and storage in SG .............................................. 37

CHAPTER 3: OPTIMAL SCHEDULING OF APPLIANCES IN SMART GRID ENVIRONMENT USING BPSO ALGORITHM ............................................................ 44

3.1 Introduction ......................................................................................................................... 44

3.2 Methodology......................................................................................................................... 46

3.3 Simulation Results and discussion ....................................................................................... 49

3.3.1 Data for simulation ........................................................................................................ 49

3.3.2 Analysis of residential appliances................................................................................... 51

3.3.3 Analysis of industrial appliances .................................................................................... 53

3.3.4 Analysis of commercial appliances ................................................................................. 55

3.3.5 Comparative analysis with GA-DSM .............................................................................. 58

3.4 Conclusions .......................................................................................................................... 59

CHAPTER 4: OPTIMAL MANAGEMENT OF HOME LOADS WITH RENEWABLE ENERGY INTEGRATION AND DEMAND RESPONSE STRATEGY ....................... 60

4.1 Introduction ......................................................................................................................... 61

vii

4.2 Load modeling and DSM implementation for multiobjective optimization ....................... 63

4.3 Microgrid Modeling ............................................................................................................. 66

4.3.1 PV system ....................................................................................................................... 67

4.3.2 Wind turbine .................................................................................................................. 67

4.3.3 Energy savings from renewables .................................................................................... 68

4.4 Results and discussion ................................................................................................. 69

4.4.1 Load profiles and scheduling of the loads ...................................................................... 70

4.4.2 Performance gain in terms of energy and cost ............................................................... 73

4.4.3 Renewable energy integration ........................................................................................ 76

4.4.4 Trade off from DSM integrated with microgrid .............................................................. 78

4.5 Conclusions .......................................................................................................................... 79

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS .................................... 81

References.......................................................................................................................... 83

Appendix ......................................................................................................................... 101

viii

List of Figures

Figure 1. 1 Smart grid conceptual model [12] ........................................................................ 2

Figure 2. 1 DSM techniques [41] ........................................................................................... 7

Figure 2. 2 Classification of DR programs [59] ................................................................... 10

Figure 3. 1 Flowchart of BPSO algorithm ........................................................................... 48

Figure 3. 2 Load curve for the residential area ..................................................................... 52

Figure 3. 3 Hourly cost curve for the residential area ........................................................... 52

Figure 3. 4 Load curve for the industrial area ...................................................................... 54

Figure 3. 5 Hourly cost curve for the industrial area ............................................................ 55

Figure 3. 6 Load curve for the commercial area ................................................................... 57

Figure 3. 7 Hourly cost curve for the commercial area ......................................................... 57

Figure 3. 8 Comparison between BPSO and GA based DSM............................................... 58

Figure 4. 1 Conceptual design of proposed microgrid model ............................................... 67

Figure 4. 2 Average hourly load profile of households during weekday ............................... 72

Figure 4. 3 Average hourly load profile of households during weekend ............................... 73

Figure 4. 4 Hourly cost curves for average load of households during weekday ................... 74

Figure 4. 5 Hourly cost curves for average load of households during weekend ................... 75

Figure 4. 6 Average power output from renewable energy sources (A) Weekday (B) Weekend

........................................................................................................................................... 77

Figure 4. 7 Hourly energy surplus and deficit for each of the households after load shifting

(A) flexible pricing weekday, (B) TOU weekday, (C) flexible pricing weekend and (D) TOU

weekend .............................................................................................................................. 78

Figure 4. 8 Monthly cost analysis under different scenarios (A) flexible pricing tariff (B)

TOU pricing tariff ............................................................................................................... 79

ix

List of Tables

Table 2. 1 Challenges of DSM implementation in the SG network and possible solutions

provided by different standards and protocols...................................................................... 11

Table 2. 2 Characteristics of various algorithms used in DSM ............................................. 19

Table 2. 3 Summary of results of the algorithms used in DSM ............................................ 24

Table 2. 4 Impacts of DSM implementation on RES integrated SG network ........................ 39

Table 3. 1 Hourly forecasted loads for different areas and electricity price [37] ................... 50

Table 3. 2 Data of residential area devices [37] ................................................................... 51

Table 3. 3 Simulation results of BPSO based load shifting [37] ........................................... 53

Table 3. 4 Data of industrial area devices [37] ..................................................................... 54

Table 3. 5 Data of commercial area devices [37] ................................................................. 56

Table 4. 1 Typical electricity tariff (average value) in Victoria, Australia ............................ 64

Table 4. 2 Appliances and power consumption pattern for households ................................ 69

Table 4. 3 Summary of the load shifting results in percentages ............................................ 76

Table A 1 Average hourly load consumption in weekday .................................................. 101

Table A 2 Average hourly load consumption in weekend .................................................. 102

x

Abbreviations

ACO Ant colony optimization

BPSO Binary particle swarm optimization

DLC Direct load control

DP Dynamic programming

DR Demand response

DSM Demand side management

EE Energy efficiency

GA Genetic algorithm

GTA Game theory algorithm

HEMS Home energy management systems

IBR Inclined block rate

LP Linear programming

MGC Micro-grid controller

MINLP Mixed-integer nonlinear programming

NLP Nonlinear programming

PAR Peak-to-average ratio

PDF Probability distribution function

PSO Particle swarm optimization

PV Photovoltaic

RES Renewable energy resources

SR Spinning reserve

TOD Time of day

TOU Time of use

WT Wind turbine

Page | 1

CHAPTER 1: INTRODUCTION

1.1 Energy and sustainability

Energy is an essential element of people’s daily life. At present, social development

depends on the usage of a sufficient amount of energy, especially electricity [1,2]. The

amount of energy consumption increases along with the rapid growth of the global population

and industrial and technological development [3]. Global energy consumption primarily

depends on the fossil fuel resources such as natural gas and coal. Thus, it is quite difficult to

meet the future energy demand due to the limited resources of fossil fuels. In addition, the

fossil fuels are responsible for the high carbon emission and global warming. According to

the International Energy Agency, approximately 70% of world’s total energy is produced

through the burning of fossil fuels, primarily coal (42%) and gas (21%) [4]. Currently,

assurance of adequate electricity supply is considered one of the most challenging tasks,

required for ensuring the continuous economic and industrial development.

Environmental sustainability and energy security are associated with the amount of

energy production and consumption. A large amount of energy resources are wasted through

the unproductive use of natural resources. Moreover, some countries, such as the United

States use oil and coal-based power plants for the production of electricity. These fossil fuels-

based power plants produce a large amount of SO2, CO2, and other greenhouse gases, all of

which are considered as a threat to the environment. Because of the growing awareness of

electricity crisis and the contribution of power generation plants to climate change, the

scientific community has started to search the alternative energy options for power

generation.

Page | 2

1.2 Smart grid

Electric network performance depends on the production of electricity and the

capability to meet consumers’ growing demand. In addition, the amount of energy

consumption certainly affects the energy distribution system. The situation becomes worse

when the distributed generation from renewable resources exceeds the penetration levels due

to the irregular and unpredictable characteristics of renewable energy [5,6]. This phenomenon

makes the operation of the grid unsafe and unreliable. Therefore, the variability in renewable

energy generation needs to be considered to meet the growing power demand and ensure grid

sustainability. The SG is an electric network with advanced sensing technologies, control

systems, and communication technologies that reflect the future of energy systems [7,8]. The

SG has been evolved with the effective distribution and supply of electricity. The main

characteristics of SG include the bidirectional flow of data and energy between the energy

provider and the customer [9]. Therefore, the SG opens the door for new prospects to supply

electricity to the consumer efficiently and dynamically. The SG has already been proven as a

convenient tool for reducing peak loads and energy costs [10]. The SG system consists of

several energy subsystems (Figure 1.1), communication, and security components [11].

Figure 1. 1 Smart grid conceptual model [12]

1.3 Demand side management

The issues associated with grid sustainability and reliability can be addressed by

DSM. The DSM is considered an essential mechanism in the energy management of SG,

Page | 3

which provides benefits to the utilities and customers. The DSM can systematically transmit

and distribute available energy to decrease carbon emissions and peak loads, as well as

allows users to choose their preferred energy type [13]. The DSM was first devised in the

year 1970 [14]. The DSM model was introduced by the electric power generation industry to

control the time of use and level of electricity demand and to analyze the profiles of

electricity loads among users. A DSM program integrated with renewable energy sources

(RESs, e.g., solar and wind), distributed micro-generators, and energy storage devices, such

as plug-in electric vehicles and batteries can provide an optimal management system by

scheduling various smart appliances and generating renewable energy [15–17]. The price of

electricity affects the usage of energy by consumers. Consumers prefer to use less electricity

if the electricity price increases. However, the implementation of the DSM in SG can easily

handle the load patterns of the electricity market as well as can analyze and reshape load

profiles. This practice reduces the peak load demand of customers, thereby improving grid

stability, sustainability, and security; additionally reduces carbon emission levels, grid

operation costs, and electricity costs [18]. Also, effective DSM activities can easily avoid the

unnecessary construction of electrical infrastructure by controlling and managing

decentralized energy resources. These activities can manage the electricity market with

consideration of power generation, transmission, and distribution.

1.4 Algorithms for solving optimization problems

The utilization of different algorithms can solve the optimization problem of DSM in

SG, which includes different technologies, such as smart meters, advanced metering

infrastructure, and communication and control technologies. The optimization problems of

SG consider the minimization of electricity costs, aggregated power consumption, and

PAR[19]. They also consider the maximization of user comfort and the efficient integration

of RESs. For example, previous studies presented different GA-based models for reducing

Page | 4

electricity costs [20–22]. Additionally, GA, ant ACO, and BPSO were used to schedule

energy consumption and evaluate the performance of HEMS controllers [23].

1.5 Problem statement and research questions

The use of fossil fuels in electricity generation is associated with a massive amount of

greenhouse emission, which is one of the major concerns of the modern world. This situation

forces electrical engineers, researches, and policymakers to find out the way for the

optimized consumption grid electricity and utilization of renewable energy resources. Based

on the knowledge gaps identified through a detailed literature review (presented in chapter 2)

following research questions are formulated.

What will be the impact of BPSO algorithm for the optimization of load scheduling

problems when compared with the genetic algorithm?

How does the flexible pricing and time of use tariff affect the load shifting and load

profile?

How does the renewable energy integrated demand response program impact on the

households’ energy consumption and cost?

1.6 Aims and objectives

The current research work aims to minimize the electricity consumption cost of the

consumer by shifting the peak load. This can be accomplished by introducing renewable

energy systems and demand response strategy. The specific objectives of the work are as

follows:

1. Scheduling of the loads employing BPSO algorithm and its comparison with other

algorithms.

Page | 5

2. Mathematical modeling of a microgrid and analysis of its effects on the electricity

cost.

3. Investigations of the effects of different tariff rates on load scheduling and energy

consumption pattern.

1.7 Outline of the thesis

Chapter 1: This chapter presents the necessity of alternative energy resources, the

significance of the smart grid on load management, the insight of DSM, and different

algorithms for solving optimization problems.

Chapter 2: This chapter discusses DSM approaches and techniques and reviews the recent

works related to the application of demand side management in the smart grid through

discussing the techniques and algorithms and their associated challenges for effective

implementation. The chapter also details the works related to the implementation of demand

side management, including the description of their operation mode, the profile of energy

production, storage and consumption, and finally, the benefit deduced by the demand side

management implementation.

Chapter 3: The focus of this chapter is to reduce the peak load demand, electricity cost, PAR

as well as to achieve substantial cost savings using BPSO algorithm based load scheduling

technique and establish a comparison with the genetic algorithm based load scheduling

technique.

Chapter 4: This chapter proposes a home energy management model, which consisted of

microgrid framework and DSM technique. The aim of this energy management model is to

minimize the monetary expenses of electricity consumption in households by shifting the

loads to the times with lower electricity prices and utilizing energy from renewables.

Chapter 5: The major conclusions of the current thesis and future research directions are

highlighted in this chapter.

Page | 6

CHAPTER 2: LITERATURE REVIEW

2.1 Introduction

The SG is considered a new opportunity to enhance the 20th century’s power grid. The

SG has gained substantial popularity due to some features, such as distributed generation,

self-healing, digital two-way communication, self-monitoring, and universal control [24,25]

The SG can adjust renewable energy generation, create smart measurement systems, and

distribute and transmit grid power by utilizing modern information and communication

technologies [26,27]. In addition, the SG can control and manage the electricity market,

construct the infrastructure, and manage the decentralized energy resources [28,29]. The

DSM supports the SG functionalities by analyzing the short and long term status of the

electricity market, determining a cost-effective option for energy supply, and modeling and

characterizing the system load [30]. However, the capacity of SG needs to be improved to

meet the growing energy demand, which requires an installation of power generation and

transmission infrastructure [31]. The development of new infrastructure will increase not only

the complexity of the SG networks but also relevant system costs. In this situation, the

efficient implementation of DSM programs in SG can overcome complexity and high

expenses by controlling and influencing energy demand. Additionally, the DSM can improve

grid sustainability by reducing the peak load demand, reshaping load profiles, and reducing

overall costs and carbon emissions. Previous studies reported the contributions of DSM on

the reduction of carbon emissions of SG. For instance, Zhang et al. reduced the carbon

emission levels in the SG environment by incorporating electric vehicles and a DR program

[32]. Ai et al. introduced a bid-scheduling DSM method in SG to motivate consumers leading

to the reduction of carbon emission levels [33]. In this case, they developed demand side

reserve scheduling based on the DR program in the SG environment. Li et al. proposed a

carbon emission flow model to investigate and evaluate carbon emission levels from different

Page | 7

parts of an SG network [34]. DSM and supply-side management jointly take part in the

energy management program. The effects of incorporating three levels of RES utilization

were also investigated for the reduction of carbon emission.

2.2 Techniques and approaches of DSM

The DSM is a growing technique for planning, implementing, and monitoring pre-

defined activities that affect consumers’ electricity utilization patterns. These activities

mainly change the time of load consumption and the utility’s total load, thereby reducing the

expected peak loads [35]. To reduce electricity costs, the DSM manipulates customers’

electricity usage patterns and produces the preferred changes in the load profiles by altering

the load shape of the power distribution network [36,37]. Essentially, the DSM helps to avoid

excess power generation by reducing peak loads leading to a reduction in operation costs

[38]. The DSM can be categorized into six major types according to the daily and seasonal

usage of electricity (Figure 2.1). These methods include load shifting, peak clipping, strategic

load growth, valley filling, strategic conservation, and flexible load shape [39,40].

DSM

Peak clipping

Valley filling

Load shifting

Strategic conservation

Strategic load growth

Flexible load shape

Figure 2. 1 DSM techniques [41]

Peak clipping: Peak clipping is a common form of the load management technique

that decreases the peak demand of an electrical network [35]. Typically, peak clipping

controls customers’ electricity consumption through DLC, which mainly explains the

system’s peak load reduction [35,37]. The DLC can be defined as a function of the

Page | 8

DSM program by which a power supplier company regulates customers’ appliances

from a distance and shifts their peak load to off-peak hours [42,43]. The proper

scheduling of DLC is considered as a favorable way to reduce operating costs and

fossil fuel usage [35,44].

Valley filling: Valley filling aims to drop the level of load difference between the

peak load and the valley load and thereby diminish load demand by filling the valley

from a curtailed load [37,39]. Valley filling is applicable when the long-run

incremental cost is less than the average price of electricity [39].

Load shifting: The load shifting technique is mostly used in the DSM program. It is

the most effective load management technique that shifts load capacity from peak

hours to off-peak hours [37,39].

Strategic conservation: Strategic conservation diminishes overall load demand

through the application of load reduction procedures by the efficient consumption of

energy. It designs and attains the desired load shape according to the planning,

distribution, and management of the network system [37,40].

Strategic load growth: Strategic load growth motivates power companies to increase

the power generation for customers [37,39,45]. It optimizes the daily response and

changes the shape of the load with respect to the large demand beyond the valley

filling technique. The activities of strategic load growth include the amplification of

the market share of loads, the economic development of service areas, and the

guaranteeing of necessary infrastructure for handling the load demand.

Flexible load shape: The flexible load shape technique mainly secures the reliability

of SG [37,39,45]. Under this technique, an electricity generation company analyzes

the load profile to identify customers with flexible loads. The customers can get

Page | 9

various incentive awards, if they control their consumption of interruptible or

curtailable load during peak periods.

Recently, governments and utility companies have focused on the implementation of

the DSM strategies that smooth the operation of electrical systems [46–48], promote and

extend energy efficiency plans and applications, and change the behavior at the customer

level or implement dynamic demand responses [49,50]. The DSM has four strategies,

namely, DR, EE, SR, and TOU [51–54]. The main emphasis of these strategies lies in the

development and use of power-saving technologies, monetary incentives, electricity prices,

and government policies to diminish peak load demands and maintain a sophisticated

synchronization between network operators and customers.

Energy efficiency: The EE is considered a modest choice with respect to the benefits

received by energy suppliers, energy consumers, and the environment [10]. The EE is

a type of technology that provides an improved and long-lasting service when the

end-use equipment is in operation. The EE programs improve the physical

infrastructure of the electrical grid for improving electricity efficiency and reducing

peak demand [55]. The characteristics of the EE programs are utility-specific and they

can store all forms of RESs. The functions of the EE programs include the change in

policies of inefficient systems, detection and replacement of misconfigured controls,

adoption of financial incentive programs, and maintenance of a level of consumer

satisfaction [56].

Time of use: A TOU pricing strategy refers to a function of fixed tariffs that divides

24 hours into several time intervals and assigns a different price for electricity

consumption in each interval [55,57]. This strategy helps to control peak period

pricing and seasonal pricing based on different prices of energy.

Page | 10

Demand response: The DR refers to a specific tariff or program which decreases or

shifts electricity usage during peak periods with respect to time-based rates or

incentive payment programs [56,58]. The network reliability of an electrical system

becomes jeopardized due to certain conditions, such as peak period network

congestion or high prices. In this situation, the DR changes the energy usage pattern

and provides an opportunity for consumers to contribute to the operation of the

electric grid [49]. The DR programs can be classified into price-based programs and

incentive-based programs, as illustrated in Figure 2.2.

Demand response programs (DRP)

Price-based programs Incentive-based programs

Time of use Real time pricing Critical peak pricing

Emergency DRP Interruptible/curtailable services Direct load control Capacity market program Demand bidding Ancillary services market

Figure 2. 2 Classification of DR programs [59]

Spinning reserve: The reserve power connected to the grid system is activated by the

system operator to maintain the balance between load and generation in case of a

sudden drop in the generation. This interruption in the power supply is caused due to

unexpected damage in generation units, incorrect load forecasting, and scheduling

[60]. Typically, the SR is classified into primary and secondary SR [55]. In primary

SR, frequency controls the active power output, whereas, in secondary SR, the

frequency and grid state is restored with additional active power [55,61].

Page | 11

2.3 Challenges of DSM implementation

The SG needs to overcome a number of difficulties related to power generation,

transmission, and distribution, as well as to the effective utilization of energy resources.

However, cybersecurity and privacy are considered as key challenges to the SG network [62].

Cyber attackers can access and misinterpret the information stored in a DSM system, which

is vulnerable to the invasion of privacy. Such information includes the software of the DSM

algorithm, load data, price signals, and users’ personal information. In addition, attackers can

easily change the load scheduling of DSM programs by introducing misinformation into the

control systems; this misinformation prompts the energy supplier to refuse to respond to the

customers’ real requests [63]. Therefore, it is necessary to provide a secure and reliable

operation between energy providers and customers. To date, a number of organizations, such

as the North American Electrical Reliability Corporation, Institute of Electrical and

Electronics Engineers, Critical Infrastructure Protection, International Society of Automation,

and US National Institute of Standards and Technology have come forward to develop

rigorous solutions for maintaining the security and privacy of SG [64,65]. Table 2.1 presents

the highlights of DSM challenges and possible solutions.

Table 2. 1 Challenges of DSM implementation in the SG network and possible solutions

provided by different standards and protocols

Challenges of DSM Possible solution

Lack of reliable

communication between

energy sources and consumers

[66].

Bluetooth or Ultra-Wide Band could be used for the

interfaces between meter and end customer devices.

IEEE 802.15.4 (ZigBee) and IEEE 802.11 (Wi-Fi) are

the technical standards which could be used for smart

meter interfaces in the home and local area network.

Page | 12


Cellular wireless for example, GPRS, UMTS or 4G

technologies could be used for the interfaces between

meters and the central system [67,68].

Lack of interoperability

among different software

applications used by electric

utilities

MultiSpeak, an industry-wide standards developed by

the National Rural Electric Cooperative Association

improves inter-operability by defining an information

model based on a programming language scheme.

Communication protocol focuses on web services and

Simple Object Access Protocol [69].

IEC-61850, an Inter-control Centre Protocol ensures the

inter-operability by specifying the definite

communication networks and systems in substations

[70].

Identification of the overall

network architecture, service

requirements, and device

capabilities as well as ensuring

the Supervisory Control and

Data Acquisition and

Automation Systems [66,69].

Recently, the European Telecommunications Standards

Institute has developed a new committee names

machine-to-machine to deal with these issues [66].

IEEE C37.1 is an IEEE Standard which deals with the

system architectures and functions in a substation as

well as covers the protocol selections, human machine

interfaces, and implementation issues. It handles the

issues related to network performance requirements,

Page | 13


reliability, maintainability, availability, security,

expandability, and changeability [71].

Lack of distributed resources

for the implementation of

demand response [69].

IEEE 1547 is a kind of standards that links the

distributed energy sources with the electric power

system in terms of interoperability, performance,

operation, testing, and safety [72].

DRBizNet is a DR business network which supports DR

program by monitoring electricity market operations as

well as creates a standard web service interface to

support DR applications. This network notifies the grid

operators automatically and accomplishes the specific

DR program for the market [68,69].

Open Automated Demand Response, a communication

protocol which supports the DR program by providing a

standardized information model [69].

Cybersecurity issues AMI-SEC Task Force developed a sets of security

requirements for AMI, which not only makes the

communication transparent between utility and industry

but also ensures the secure communication. These can

be achieved by the transmission of security parameters,

cryptographic key establishment, and management [73].

The North American Electric Reliability Corporation

provides a set of security requirements including critical

Page | 14


cyber asset identification, training in cybersecurity,

security management controls, electronic security

perimeters, incident reporting, and response planning,

information protection, recovery plans for critical cyber

assets and physical security [69].

To manages the power system and secure the

information exchange, IEC 62351 provides

cybersecurity requirements. They provide end to end

information security by algorithms and secure

manufacturing message specification. This protocol also

focuses on the security policies, access control, and key

management of the system [74].

According to the researchers, fairness is another main concern for load management

[75–79]. In order to motivate the customers to shift their load, they need to be assured that

they will pay a minimal amount for their electricity consumption or receive financial

incentives [77]. Therefore, a suitable fairness condition should be maintained to assess the

fairness of the algorithms that can help to choose an appropriate DSM program in practice.

Energy system efficiency depends on the optimization of different communication protocols,

SG applications, and control infrastructure [9]. However, due to the nature of distributed

control problems and the interdependency of different domains (i.e., power, communication,

and control), it is entirely difficult to develop an advanced communication infrastructure.

Additionally, the lack of standardization and interoperability among DSM entities inhibits the

Page | 15

possible integration of advanced applications of smart meters, smart devices, and RESs.[80]

Some DSM objectives, such as continuous interoperability, increased safety of new products

and systems, robust information security, compacted set of protocols, and information

exchange can be easily achieved through the standardization and interoperability of SG [81].

Inaccurate time measurements and automated analysis, fast control messaging, poor

visibility, slow response times, system handling under contingency, and lack of situational

awareness are also bottlenecks for effective DSM implementation [80,82,83]. Also, rising

population and demand for energy, energy storage problems, global climate change, decrease

in fossil fuel sources, equipment failures, capacity limitations of electricity generation, and

flexibility of problems are prime concerns, according to the researchers [84].

2.4 Progress of DSM models and applications of algorithms

The existing literature indicates that various types of algorithms, including single and

hybrid ones, have been developed and implemented to solve the optimization problems of the

DSM of SG. The PSO, GA, GTA, ACO, LP, NLP, and DP are the most widely studied

algorithms in the field of DSM. Recently, hybrid algorithms have gained remarkable attention

as promising methods. Table 2.2 shows the characteristics and user-defined parameters of

various algorithms used in DSM. This section provides an overview of the research and

advancement of algorithms used in DSM.

The GA has been applied to many optimization problems to achieve the desired

objectives of DSM. Table 2.3 summarizes the algorithms used for solving the DSM

optimization problems in SG. The reviews related to the application of the GA in DSM are as

follows. Logenthiran et al. proposed a day-ahead load shifting technique for the DSM of SG

[37]. They used a heuristic-based evolutionary algorithm to solve a minimization problem

and shift the load from peak hours to off-peak hours. The simulations were carried out in

residential, commercial, and industrial areas with a variety of loads. In this work, the authors

Page | 16

considered the central controller of the SG to control the DSM technique. However, if the

controller becomes disabled for any reason, this whole process will stop functioning.

Compared with other areas, residential areas showed the highest number of devices available

for control. However, the amount of reduction in operating cost was not as expected because

no incentive scheme or financial reward plan was considered to motivate the customers to

shift their load. In another study, a GA-based DSM method was proposed to solve the

objective function [85]. In this case, the method considered only an industrial load. The

authors used a load shifting technique to reduce customers’ inconvenience, energy generation

cost, and total electricity cost. Similarly, Bharathi et al. applied a heuristic-based GA to

model the DSM [86]. The proposed DSM model reshaped the load patterns and reduced

energy usage in industrial, commercial, and residential areas by using a suitable load shifting

technique. Arabali et al. introduced a GA-based SG strategy for shifting residential cooling

loads to match renewable energy production [22]. The authors recommended using the

developed approach in heating, ventilation, and air conditioning loads. Yao et al. developed a

DSM model based on a modified GA, named iterative deepening GA, to optimize the

scheduling of DLC approaches and minimize the revenue loss of electricity generation

companies [44]. They considered only the air conditioning load, which has a low impact.

Therefore, various types of loads need to be considered for further justifying the iterative

deepening GA.

The PSO algorithm-based techniques have been widely implemented by researchers

to solve energy management problems. The PSO algorithm is considered as significantly

effective in solving various optimization problems [87]. The PSO-based techniques are

reviewed in the recent study. Logenthiran et al. modeled a day-ahead load scheduling

technique that incorporates the PSO algorithm based on the customers’ inputs and forecasted

hourly electricity rates [88]. In this study, the authors considered the shiftable and non-

Page | 17

shiftable loads controlled by a central controller of the SG. The simulation was carried out in

residential, commercial, and industrial areas, and the results revealed a reduction in PAR and

an increase in electricity cost savings. However, no incentive scheme or reward plan was

applied to compensate the customers for giving up their comfort zone and shifting their load

from peak hours to off-peak hours. Nayak also developed a PSO-based DSM strategy that

considers a load shifting technique mainly for residential loads [89]. The methodology

comprises mostly population-based heuristic optimization techniques, which are used to solve

the scheduling problems and provide global optimum solutions.

Page | 19

Table 2. 2 Characteristics of various algorithms used in DSM

Algorithm type Algorithm

name

Mechanism User-defined

parameters

Characteristics Ref.

Metaheuristic and

evolutionary

algorithm

GA Inspired by the

mechanism of

natural

selection.

The size of the

population of

solutions, the number

of parents, the

probability of

crossover, the

probability of mutation

and the termination

criterion.

Genes of chromosome represent the decision

variable. This variable contains binary, continuous

or discrete values.

Genetic operators are responsible for the creation

of new solutions.

Individual chromosome provides a possible

solution and parents provide an old solution while

a new solution is provided by offspring. Elite

provides the best solution.

Population diversity and selective pressure affect

the search method.

Correction of convergence depends on the

selection of a good termination criterion and

[90–92]

Page | 20


name


parameters


optimum selective pressure.

PSO Inspired by the

social behavior

of birds flocks.

Size of the population

of solutions, the value

of the initial inertia

weight, the final value

of the inertia weight,

and the termination

criterion.

The decision variable is represented by the particle

position in each dimension.

The solution of the optimization problem is found

by the position of the particle where the position is

updated to find a new solution.

Fitness function is measured by the distance

between particle and food.

A number of iterations, selection of good

termination criteria, the improvement of the

objective function, and the run time of the

algorithm determine the confection of

convergence.

[90,93–

96]

Page | 21


name


parameters


ACO Inspired by the

collective and

searching

behavior of ant

species.

The size of the

population of

solutions, the

evaporation rate, the

control parameters of

pheromone, heuristic

information, and the

termination criterion.

Decision variables are represented by the path of

an ant.

In the case of an optimization problem, a possible

solution is determined by the tour of an ant from

nest to food.

Process of generating new solutions is accelerated

by the information-based stochastic mechanism.

ACO allocates desirability to the decision space

according to the fitness value of a solution.

Correction of convergence depends on the number

of iterations, selection of good termination criteria,

the incremental improvement of the objective

function, and the run time of the algorithm.

[90,97–

100]

Page | 22


name


parameters


Classical method LP Mathematical

programming

method where

the objective

function is

linear.

The collection of

coefficients with

respect to decision

variable, constraint,

the upper bound is the

parameter of the LP

method.

Objective functions are correspondents to a

restricted set of constraint.

It has a feasible solution and region.

The optimal solution can be found.

Multiplicity in solutions.

[101–104]

NLP Mathematical

programming

with respect to

the nonlinear

objective

function.

Parameters are defined

based on the problems

Converting a complex problem into an easy

problem.

Solving the sequence of sub-problems.

Solving of sub-problems are involved with the

unconstrained minimization function.

The optimal solution can be found.

[105–108]

Page | 23


name


parameters


DP Multistage

nature of

optimization

method.

There is no specific

parameter. Each

problem has its own

parameter.

Representing the multistage decision process.

For each stage, a policy decision is requested.

Solving multivariable optimization problem.

In order to determine the optimal solution for the

problem, the solution method is categorized.

Recursive relation is used to optimize the solution

procedure.

[108,109]

Page | 24

Table 2. 3 Summary of results of the algorithms used in DSM

Algorithm Methodology Reduction of

PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)

Fairness Software Remarks Ref.

Genetic

algorithm

Heuristic-

based load

shifting

technique

14.2–18.3

Customers’

savings: 5–10

Company’s

savings:15–20

× 5–10 × Real-time

simulation

The algorithm

converged

well and

globally.

[37]

Multi-

objective

particle swarm

optimization

method based

on the fuzzy

technique

Probabilistic

model-based

on incentive

payment

demand

response

programs

× × 14 21 × MATLAB Pollution

emission

factor was

considered.

[110]

Page | 25


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


Game theory Energy

consumption

scheduling

approach

38.1 37.8 × × × MATLAB The algorithm

converged

locally.

[111]

Game theory Distributed

energy storage

planning

31.5 22.43 × × √ MATLAB The algorithm

converged

globally with

minimum

information

exchange.

[112]

Particle swarm

optimization

Load shifting

technique

43 18 × × × MATLAB Algorithm

converged

globally and

[88]

Page | 26


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


required less

time interval.

Non-stationary

DSM

algorithm

based on a

repeated game

framework

Incentive

compatible

10 50 (including

discomfort and

billing cost)

× × √ Real-time

simulation

A simulation

was carried

out in the

homogeneous

and

heterogeneous

situations. The

algorithm

converged

globally.

[42]

Page | 27


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


Vickrey–

Clarke–Groov

Vickrey–

Clarke–Groov

pricing method

19.3 37.8 × × √ MATLAB The algorithm

converged

globally with

50 users.

Shifting and

executing the

algorithm

required a

certain amount

of time.

[113]

Distributed

algorithm

DSM scheme

based on time-

varying pricing

64.76 × × × × MATLAB

optimization

solver

The residential

load was

considered,

[114]

Page | 28


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


Mosek and a

simulation was

carried out on

35 appliances.

Game theory Autonomous

and distributed

DSM scheme

17 19.6 × × √ MATLAB The algorithm

converged

locally and

required

minimal

execution

time.

[115]

Non-linear

mixed-integer

Dispatching

model with DR

× Reduced × × × MATLAB The optimal

dispatching

[116]

Page | 29


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


linear

programming

model was

unable to

handle large

loads.

Any colony

optimization

Congestion

management

method

× Reduced × × × MATLAB The

integration of

RESs was not

considered.

[117]

Distributed

algorithm

Autonomous

energy

scheduling

scheme

24 21 × × × MATLAB The algorithm

converged

locally.

[118]

Page | 30


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


Sequential

Gauss–Siedel

algorithm

Parallel

autonomous

optimization

scheme with

DR framework

19.71 5.53 × × × Real-time

simulation

The algorithm

converged

locally.

[57]

Game theory Energy

consumption

and storage

optimization

method

40 19 × × × Real-time

simulation

The algorithm

converged in

parallel. The

algorithm was

able to handle

a large number

of users and

simultaneously

[119]

Page | 31


PAR (%)

Minimization of

energy cost (%)

Reduction

of carbon

emission

Reduction

of

operating

cost (%)


update their

strategies.

Genetic

algorithm

Load shifting

technique

23.84 × × × × MATLAB The algorithm

performed

well.

[86]

×: Not considered; √: Considered

Page | 32

The multi-objective PSO algorithm was studied by Aghajani et al. for reducing the

operating cost and emission with the integration of renewable sources in the micro-grid [110].

They recommended stochastic programming that focused on probability density functions

and was integrated with the DR model to optimize the performance of smart micro-grids.

However, the uncertain behavior of predicted power generation from wind turbines and solar

cells remarkably affects the operation cost. Therefore, some existing studies focused on the

implementation of BPSO in DSM, which is a modified form of the PSO algorithm. Kennedy

and Eberhard first employed the BPSO algorithm to schedule interruptible loads and solved a

multi-objective optimization problem [120]. Pedrasa et al. also suggested the implementation

of BPSO to optimize DSM problems [121]. A realistic scheduling mechanism based on the

BPSO for SG was suggested by Mahmood et al.[122]. They found the BPSO algorithm to be

an effective algorithm for reducing electricity costs. In this case, the appliances were

categorized according to the respective constraints and effective time of usage for increasing

the appliances’ utility. Zhou et al. proposed a real-time optimal appliance usage approach to

maintain energy usage based on the BPSO algorithm [123]. The method smoothed the peak

shaving, valley filling, and demand curve as well as reduced energy usage with the assistance

of customers and energy suppliers. Zhou and Xu also applied the BPSO algorithm to solve

the cost function of electric vehicle users, SG, and power sources [124]. The simulations for

load shifting, energy-saving, and energy supply efficiency were carried out in the MATLAB

platform.

Extensive research based on the implementation of ACO algorithm has been

performed to handle energy management optimization problems. Dethlefs et al. used a

distributed ACO-based self-optimization method for producing a day-ahead schedule [125].

In their work, only shiftable loads were considered to reduce the external purchase of energy.

The algorithm optimized the load of distributed consumers and the power generated from a

Page | 33

wind power plant. In this case, almost 10% of the power rating was used to control the

residential shiftable appliances, and the algorithm mainly adjusted the mean generation and

demand. An efficient DSM model based on the ACO was presented by Rahim et al. to control

residential energy [126]. The model reduced the peak load, PAR, and electricity costs,

considering the customers’ satisfaction levels. They also used a TOU tariff model with an

inclined block rate to avoid the peak load and complexity in the estimation of electricity bills.

Hazra et al. explored an efficient method for handling the load congestion problems in SG in

an economical way [127]. The problem was solved by using DR, ACO, and fuzzy techniques.

Their findings suggested that the method was able to reduce the electricity cost and fulfill

customers’ satisfaction through the scheduling of different generation resources. In another

study, the authors analyzed the load congestion management problems to control model cost

[117]. The problem of real-time congestion management was developed as an NLP problem.

In this study, the ACO algorithm provided a feasible solution for the problem and minimized

the electricity cost. Okonta et al. proposed an ACO-based load scheduling algorithm for a

smart home [128]. The researchers mainly focused on the total electricity bill, TOU, and the

overall increase in quality of life with the incorporation of the optimal utilization of

integrated RESs. An automated load manager based on the ACO algorithm and an interactive

web interface were used for DLC and energy management to allow users to access their home

appliances from remote areas via the internet.

Recently, the GTA algorithms have gained remarkable popularity in solving the DSM

optimization problems because of their capability of solving distributed system problems. In

addition, designing an algorithm using game theory is relatively easy [129]. Hung Khanh

Nguyen et al. applied a non-cooperative game theory model to formulate the DSM problems

[31]. The model reduced the peak demand, PAR, and total energy cost. In this case, the GTA

was not able to converge for the optimal solution in the centralized design. Nevertheless, the

Page | 34

algorithm minimized the PAR, close to the optimal solution of the centralized design. Here,

the impacts of large numbers of users with battery on system performance were not

investigated. The large numbers of users integrated with battery can influence the aggregate

cost and PAR reduction. Song et al. [42] developed an optimum non-stationary DSM model

based on a repeated game model, which was mainly an incentive-compatible model. This

model was designed to allow consumers to choose their daily consumption patterns

independently without affecting their habits, choices, and wants. Under this strategy, an

active set of consumers was selected based on the historical energy consumption patterns, and

the billing cost compensation was considered to motivate the consumers. Wang et al. studied

a DSM model integrated with cognitive radio technology based on the proficient and

trustworthy communication infrastructure in SG [112]. This model suggested a cost function

focusing on customers’ preference with a balanced payment model consisting of billing,

electricity generation, and discomfort costs. This DSM system was designed to allow users to

select an appropriate size of storage units for balancing the costs. In this study, the GTA was

applied to optimize the distributed planning storage method. The results indicated the

reduction of PAR, total energy cost, customers’ daily payment, and energy consumption. This

work also proved that the application of cognitive radio technology can effectively reduce

energy consumption in the SG communication networks. However, customers’ privacy was

not guaranteed in this case. Deng et al. analyzed the residential energy consumption

scheduling problem and formulated a couple of constrained game with respect to interactions

among customers [130]. They applied a real-time pricing approach that shifted the peak

demand to the off-peak hours to balance the energy demand. A GTA-based autonomous and

distributed DSM scheme was proposed by Mohsenian–Rad et al. [131], who focused on the

scheduling of energy consumption with the consideration of residential loads. The proposed

technique was based on incentives and thus reduced the peak load, total energy costs, and

Page | 35

customers’ daily electricity charges. In this investigation, the authors employed a new energy

cost function; however, the estimation of energy consumption and simulation was time

consuming. In another research, Nguyen et al. proposed a smart power system with an energy

storage device based on the GTA [119]. The objective of the study was to diminish the square

Euclidean distance between the instantaneous and regular electricity demands of the energy

system. The power consumption was scheduled using the energy cost-sharing model, and the

loads were synchronized by a principal controller.

Many researchers have applied the LP method to solve DSM optimization problems

and their results are summarized in the current work. Sheblt studied a load management

scheduling program using the LP method [52]. A DLC scheme was used to schedule

appliances and to increase the profit of energy suppliers according to cost or market price

function. Kurucz et al. also applied the LP model for scheduling the loads under control

periods to minimize the system peak load [132]. The LP model considered the residential,

commercial, and industrial loads to bring a specific number of customers under the model

and determine long-term and short-term control scheduling strategies. An integer LP-based

load scheduling mechanism was proposed by Zhu et al. for the DSM of SG [133]. With the

aim of reducing peak hour loads, they used the proposed mechanism to schedule home

appliances along with the optimal power and operation time according to customers’

preferences. Martins et al. proposed a multi-objective LP model to increase the power

generation capacity [134]. The total extension cost, the environmental impact associated with

energy output, and the environmental effects associated with the installed power capacity

were considered in the investigation. In order to solve the objective function, they also

considered five constraints, such as the reliability of the supply system, the availability of

generation units, and the capacity of the generation group equivalent to the DSM.

Page | 36

Many researchers have used the NLP method in DSM because it improves cost

functions and generates satisfactory results [135]. Shaaban et al. used a MINLP method for

scheduling batteries and shiftable and adjustable loads [136]. The implementation of this

energy management technique was able to reduce the operating cost under an SG network.

Wang et al. also used the MINLP method to optimize the optimal dispatching model of a

smart HEMS with distributed energy resources and intelligent domestic appliances [116].

This method reduced electricity costs and aggregated power consumption. However, the

MINLP was not able to handle many appliances because of the unpredictable, impulsive,

non-linear, and complex energy consumption patterns of consumers. Considering DR, Helal

et al. proposed a mixed-integer NLP-based energy management model to optimally schedule

the different generation technologies of AC/DC hybrid micro-grids in islands [137]. They

also suggested that the system depends on a MGC, which ensures the proper usage of energy

with minimum operating costs by controlling user appliances and water desalination units.

An optimum schedule with the minimum cost was achieved by formulating the scheduling

problems as MINLP problems.

Thus far, this model has been used in several studies to solve the DSM optimization

problems of SG. Chu et al. proposed a dynamic programming-based optimization algorithm

to determine the scheduling schemes of DLC [138]. The authors considered an air conditioner

in a commercial building for load scheduling. The method reduced the peak load according to

customers’ discomfort level. In addition, this method was found to be effective in scheduling

other appliances by determining a target load level and controlling load usage to reduce the

peak load and electricity generation costs [29]. Primarily, an analytic dynamic programming

model was used to schedule some of the appliances of residential loads. In this study, the

performance of the algorithms and the control periods of appliances were investigated in five

cases; however, the satisfaction levels of customers were not considered. Hsu et al.

Page | 37

introduced a dynamic programming-based optimization method for reducing the system’s

energy generation cost for the DLC dispatch [139]. Therefore, the DLC strategies were

integrated with a unit commitment problem, and a DP method was developed to solve

dispatch DLC and the unit commitment problem. To reduce the electricity demand in the SG

environment, Reka and Ramesh proposed a DP model with a cloud computing framework

[140], which created a small energy hub with customers and displayed the customers’

participation in DSM programs.

2.5 Integration of renewable energy sources and storage in SG

Renewable energy resource (RES) like solar, wind and their hybrid system has

become a popular means of energy supply. However, the integration of RES with the SG has

been gone through a complicated situation because of mixing a number of energy resources

and their intermittent behavior [141]. Therefore, the DSM has been incorporated in RES

integrated SG to handle the fluctuation of electricity price, the mismatch between renewable

energy generation and load demand as well as control of power transactions [142,143]. The

DSM has a significant impact on the RES and energy storage unit as the implementation of

DSM in SG shifts the loads from peak hours to off-peak hours, which allows storage unit to

store the excess power produced from the RES or in the time when grid electricity is cheap.

The stored energy can be used in the future when the energy supply is shortage and peak

periods, which can add economic value in the grid. Table 2.4 shows the summary of DSM

implementation in the RES integrated smart grid network. Quiggin et al. modeled a

residential microgrid integrated with renewable generation technologies, energy storage, and

DR systems [144]. The implementation of the DR program in this model was able to reduce

the peak demand fluctuations by 16% and optimize the energy balance between supply and

demand. Dietrich et al. analyzed the effect of the DR program on the wind energy demand

profile in terms of cost reduction in the system [145]. Incorporation of the DSM program

Page | 38

reduced the number of generation units and flattened the electricity production curve.

Aghajani et al. showed that the utilization of the DSM method reduced the effect of

uncertainty, which was caused during energy generation from solar cells and wind turbine

[110]. Çiçek and Deliç reported that the DSM method was able to achieve a steadier pattern

of social welfare, which was measured in terms of customers’ utility and energy generation

cost [146]. This method maintained the balance in the integration of the wind farm to the grid

and dealt with the issues of energy fluctuations and economic risk. Amrollahi and Bathaee

showed that the DR program maintained the energy distribution in such a way that it reduced

the required number of batteries, inverters, and PV cells as well as reduce the total cost [147].

Besides, the implementation of DSM improved the load factor and correlation factor by 57.9

and 36.8%, respectively. Wang et al. formulated a hybrid RES with DR program and applied

to a single-family residential home [148]. The implementation of DR scheme met the

consumers’ electricity demand by utilizing the available power from PV panels, wind turbine,

diesel generator, and batteries. This method improved the system efficiency when compared

with that of the traditional method. Behboodi et al. applied DSM technique with the

integration of RES to solve the multi-area electricity resource allocation problems [149]. This

method offered an uniform electricity price by maintaining a steady-state among energy

generation, transmission, and load constraints. The researchers also described an innovative

approach to solve multi-area electricity resource allocation problems considering both

intermittent renewables and DR. The method determined the hourly inter-area export/import

set that maximizes the inter-connection surplus satisfying the transmission, generation, and

load constraints.

Page | 39

Table 2. 4 Impacts of DSM implementation on RES integrated SG network

Integrated system DSM

method/technique

Operation mode Supervisory

control

Outcome Reference

Photovoltaic-battery

hybrid system

DR program with

Model predictive

control method

Grid-connected Centralized Minimized the electricity bill on

the customer side.

Maximized the use of solar

energy and battery storage.

[150]

Industrial microgrid with

wind turbine and energy

storage unit

DR scheme Grid-connected Centralized Wind turbine reduced the carbon

emission by 88% and DSM

produced 30% more reduction.

Overall electricity cost reduced by

73%

[151]

Residential microgrid

with photovoltaic panel,

wind turbine, and energy

storage unit

DR scheme with linear

programming method

Grid-connected Decentralized Energy demand was reduced by

16%.

During all hours of operation, the

reduction of CO2 emission along

[144]

Page | 40


method/technique


control

Outcome Reference

with the associated energy usage

was 10%.

During the hours of operation, the

amount of renewable supply was

reduced by 74%.

Microgrid system with

photovoltaic panels, wind

turbine, diesel generator,

battery bank, and water

supply system

DSM mechanism

along with artificial

neural network

Grid-connected Decentralized Minimized the operation cost by

3.06%.

[152]

Household dotted

with photovoltaic systems

Load scheduling

method based on

online event-triggered

energy management

Grid-connected Centralized Reduced the electricity bill as

well as ensured the user comfort

level.

[153]

Page | 41


method/technique


control

Outcome Reference

algorithm

SG network with

renewable distributed

generators

DR scheme with

parallel autonomous

optimization

Grid-connected Centralized Reduced electricity generation

costs and electricity bills.

[154]


micro turbines, wind

turbine, fuel cells,

photovoltaic panels,

storage devices and a

group of radial load

feeders

DR scheme Grid-connected Centralized The peak load was shaved from

the grid tie-line.

Achieved optimal scheduling of

batteries and diesel generators.

[155]

Microgrid with renewable

generators and energy

storage

DR scheme Isolated Centralized Achieved an optimal power

generation and peak load

dispatch.

[156]

Page | 42


method/technique


control

Outcome Reference

SG network with high

wind penetration

DR scheme Isolated Centralized Achieved 30% cost savings.

More than 56% of demand was

shifted.

[145]

SG network with the

energy storage device

Load scheduling with

game theory algorithm

Grid-connected Centralized Reduced peak load as well as

energy payment for the

consumers.

[119]

Microgrid network with

wind turbine and solar

cell

DR scheme Grid-connected Decentralized Reduced operational cost and

carbon emission.

[110]

SG network with wind

farm

DR scheme Grid-connected Centralized Achieved optimal scheduling of

energy production and

consumption for 24 hour.

[146]


photovoltaic system,

DR method with

mixed-integer linear

Isolated Decentralized Operational cost and peak load

were reduced by 17.2 and 36.8%,

[147]

Page | 43


method/technique


control

Outcome Reference

wind turbine, and battery programming respectively.

Energy system with

photovoltaic panels, wind

turbine, diesel generators,

and batteries

DR scheme Grid-connected Centralized Reduced operational cost and

environmental cost.

[148]

SG network with

photovoltaic system

DR scheme Grid-connected Decentralized Load factor is analyzed and

increased during a year.

[157]

Page | 44

CHAPTER 3: OPTIMAL SCHEDULING OF APPLIANCES IN SMART GRID ENVIRONMENT USING BPSO ALGORITHM

Abstract

In this work, BPSO algorithm is used for DSM implementation in the SG

environment. Load shifting technique is applied in the residential and industrial area and

shifted the load from peak hours to off-peak hours. Load shifting technique is mathematically

formulated and implemented as a minimization form. In this work, it has been clearly shown

that BPSO based load shifting method can be able to handle a large number of devices of

various types compared to the traditional DSM method. The focus of this work is to reduce

the peak load demand, electricity cost, PAR as well as to achieve substantial cost savings.

BPSO based load shifting method shows a better result in terms of peak load reduction when

compared to GA based DSM.

Keywords: Smart grid, DSM, Algorithms, BPSO, Load shifting.

3.1 Introduction

The electricity demand is expected to increase to almost twice the current demand by

the year 2020 because of the rapid electricity consumption as a consequence of the quick

movement of globalization and industrialization [3,158]. Therefore, effective utilization and

distribution of power supply are necessary to maintain continuous economic and industrial

development. SG brings the highest opportunities to handle the future energy system. SG is

an electric grid network with advanced sensing technologies, control methodologies, and

communication technologies, which provides bi-directional communication between the

consumers and electricity suppliers [7,8].

DSM opens the new door for the efficient supply of electricity by implementing

policies and measures for energy consumption [159]. In order to achieve an optimistic power

Page | 45

consumption curve, DSM is implemented directly or indirectly by the utility companies.

Moreover, an efficient DSM with SG helps to achieve an optimistic utilization and

distribution of electricity by varying the price tariff of electricity between the peak and off-

peak hours [86]. Generally, consumers have to spend a lot of money because of the high price

of electricity. However, the integration of DSM with SG reshapes the load profile and

provides the desired load curve to the utility companies. DSM manages the accessible

electricity in the grid from various perspectives, for example, residential, commercial, and

industrial. This situation leads to the reduction of peak load demand, electricity cost of the

customer as well as improves the grid stability, sustainability, and security.

So far, a number of algorithms, such as dynamic programming, linear programming,

and heuristic evolutionary algorithm have been employed for solving the DSM problems

[37]. For example, Kurucz et al. and Shaaban et al. proposed a linear programming algorithm

for scheduling the load and minimizing the peak load demand [132,136]. Logenthiran et al.

proposed a GA based load shifting technique. They managed the load demand in the case of

residential, commercial, and industrial areas. They were able to achieve a substantial cost

savings in terms of peak load reduction and electricity cost [37]. To minimize the system

production cost, an optimization technique based on dynamic programming was also

investigated [160]. In this paper, the DLC technique was implemented, which gave access

and permission to utilities to directly control the portion of the customers' load. Anvari-

Moghaddam et al. developed a multi-objective mixed integer nonlinear programming model

for optimal usage of electricity in a smart home [161]. They mainly focussed on the

electricity savings and comfortable lifestyle in terms of reduction of residential electricity

usage and utility bills. A multi-objective evolutionary algorithm is proposed by Muralitharan

et al. [162]. The model was presented for the DSM application for obtaining the energy cost

savings and able to minimize the appliances’ waiting time. To reduce the peak load demand,

Page | 46

Pallotti et al. proposed a GA based optimization problem. This method was able to achieve an

optimal planning of energy consumption for 246 smart homes which focuses on the energy

cost minimization and user satisfaction [163]. Kinhekar et al. presented a multi-objective

DSM problem formulation and solution based on the integer GA algorithm [164]. The DSM

solution is provided based on the forecasted load data, pool market price, and TOD tariffs.

Load shifting technique used to schedule shiftable appliances for both commercial and

industrial consumers area. To handle the home load, Rastegar et al. proposed a DR

mechanism for residential households [165]. To obtain a optimal scheduling for household

devices, the DR program was applied under the pricing scheme of TOU and IBR. This work

reduced the electricity bills and maximized the user comfort based on controllable and

uncontrollable appliances. Maximum user comfort level was achieved by focusing on the

functional hours of appliances. Multi-objective optimization problem based on MINLP was

proposed by Shirazi and Jadid [165] to obtain an optimal load scheduling for the residential

sector. The simulation was carried out for both winter and summer day in order to minimize

the energy consumption cost and maximize the customers’ comfort. Electrical and thermal

appliances were both considered for shifting the load from peak to off-peak hour. However,

these algorithms cannot be able to handle the large numbers of various types of devices

because of their system specific nature. The aim of this work is to solve a minimization

problem by shifting the load from peak hours to off-peak hours employing the BPSO

algorithm and establish a comparison with GA. In this case, residential, industrial and

commercial appliances are considered for the investigation.

3.2 Methodology

This research work presents a BPSO based load shifting technique for DSM. Here,

day-ahead load shifting technique is applied based on the forecasted load and electricity

price, where controllable and uncontrollable loads are considered for the optimization

Page | 47

problem. For all the cases (i.e., residential, commercial, and industrial area), only the

controllable loads were shifted based on the electricity cost and user preference. In this

method, residential, commercial, and industrial appliances are considered where each device

has different energy consumption. Load shifting technique is applied in a minimization form

and presented as follows.

Minimize, ∑ (𝑃(𝑡) − 𝑂(𝑡))2𝑀𝑡=1

where, 𝑃(𝑡)= actual consumption at time t.

𝑂(𝑡) = objective curve at time t.

P(t) can be expressed by the following equation.

P(t)=F(t)+C(t)-D(t) (3.1)

where, F(t)= forecasted loads at time t.

C(t)= loads connected at time t.

D(t)= loads disconnected at time t.

The PAR is calculated using the following equation.

PAR= 𝑙𝑜𝑎𝑑𝑝𝑒𝑎𝑘𝑙𝑜𝑎𝑑𝑚𝑒𝑎𝑛

(3.2)

In BPSO algorithm, the number of hours in a day was represented by a particle and

the particle was represented by a row vector with m variables. BPSO algorithm is a modified

version of the PSO algorithm as shown in Figure 3.1. In the case of BPSO, each particle of

the population has to decide whether the decision is true or false. This true/false decision was

taken place based on the binary value 0 and 1. If the decision is true, the value is 1 and if it is

false, the value is 0. For this reason, binary values are arranged into a particle in a search

space where the optimization technique is applied.

Page | 48

Start

Find and update pbest and gbest

End

Termination criteria

Yes

No

Evaluate the fitness of particles

Calculate and update the velocity of particles

Initialise particles and velocity vectors

Calculate and update the position of particles

Show gbest

Figure 3. 1 Flowchart of BPSO algorithm

The probability of the decision is true and false can be defined by the following equation

[166]:

𝑃(𝑥𝑖𝑘 = 1) = 𝑓(𝑥𝑖𝑘(𝑡 − 1), 𝑉𝑖𝑘(𝑡 − 1), 𝑝𝑖𝑘,𝑝𝑚𝑘) (3.3)

Page | 49

Here, 𝑃(𝑥𝑖𝑘 = 1) is the probability where the ith individual chooses 1 for the kth bit in the

string, which depends on 𝑝𝑖𝑘,𝑝𝑚𝑘

𝑓(𝑥𝑖𝑘(𝑡 − 1) is the function of a previous position of a bit

𝑉𝑖𝑘 defines the individual tendency to select 1 or 0.

𝑉𝑖𝑘 can be mathematically expressed by the sigmoidal function

S(𝑉𝑖𝑘) =1

1+exp (−𝑉𝑖𝑘 ) (3.4)

𝑥𝑖𝑘(𝑡) = 1 𝑤ℎ𝑒𝑛 𝑝𝑖𝑘, < 𝑆(𝑉𝑖𝑘) (3.5)

𝑥𝑖𝑘(𝑡) = 0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒 (3.6)

𝑝𝑖𝑘 is a uniformly distributed random numbers within the range [0, 1].

3.3 Simulation Results and discussion

3.3.1 Data for simulation

Table 3.1 shows the hourly forecasted loads for residential, commercial, and industrial

areas with electricity prices [37]. These loads include both shiftable and nonshiftable loads

with different time steps. With the aim of comparing the BPSO based load shifting technique

with GA based technique, the data for the residential, industrial, and commercial appliances

were taken from the previous study [86]. This makes the comparison results more reliable.

The load shifting simulation was carried out for both residential and industrial loads in the

MATLAB software platform.

Page | 50

Table 3. 1 Hourly forecasted loads for different areas and electricity price [37]

Time (h) Price (ct/kWh) Forecasted load (kWh)

Residential Industrial Commercial

1 8.11 412.3 876.6 375.2

2 8.25 364.7 827.9 375.2

3 8.1 348.8 730.5 404

4 8.14 269.6 730.5 432.9

5 8.13 269.6 779.2 432.9

6 8.34 412.3 1120.1 432.9

7 9.35 539.1 1509.7 663.8

8 12 729.4 2045.5 923.5

9 9.19 713.5 2435.1 1154.4

10 12.27 713.5 2629.9 1443

11 20.69 808.7 2727.3 1558.4

12 26.82 824.5 2435.1 1673.9

13 27.35 761.1 2678.6 1673.9

14 13.81 745.2 2678.6 1673.9

15 17.31 681.8 2629.9 1587.3

16 16.42 666 2532.5 1558.4

17 9.83 951.4 2094.2 1673.9

18 8.63 1220.9 1704.5 1818.2

19 8.87 1331.9 1509.7 1500.7

20 8.35 1363.6 1363.6 1298.7

21 16.44 1252.6 1314.9 1096.7

22 16.19 1046.5 1120.1 923.5

23 8.87 761.1 1022.7 577.2

24 8.65 475.7 974 404

Page | 51

3.3.2 Analysis of residential appliances

In the case of residential loads, the simulation was carried out in a total of 1547

devices of 7 types, as shown in Table 3.2. Compared to industrial and commercial areas,

residential appliances have lower electricity consumption ratings and short periods of

operation. It can be seen from Figure 3.2, the implementation of DSM shifts the loads from

peak hours to off-peak hours and provides a load curve close to the objective curve. Without

DSM strategy, the value of peak demand was about 1363.6 kWh at 20th hour and reduced to

967.1 kWh when DSM was implemented. With DSM, the peak load shifted to the first hour

with a value of 1018 kWh. The hourly cost curve (Figure 3.3) depicts that the electricity cost

reduced significantly in the peak hours with DSM and remained nearly constant from the

third hour. Table 3.3 presents the comparison between the results with DSM strategy and

without DSM strategy in terms of peak demand, PAR, and electricity cost. As can be seen,

the DSM strategy reduces the PAR, peak demand, and total electricity cost by 25.41, 25.35,

and 16.87%, respectively.

Table 3. 2 Data of residential area devices [37]

Load types Power consumption of load (kWh) Number of

devices 1st

hour

2nd

hour

3rd

hour

4th

hour

5th

hour

6th

hour

Central AC 5 5 5 5 5 5 60

Well pump 8 8 8 0 0 8 430

Hair dryer 2.4 2.4 2.4 2.4 0 0 158

Dish washer 1.2 1.2 1.2 1.2 0 0 290

Vacuum

cleaner

1.2 1.2 1.2 0 0 0 248

Laptop 0.75 0.75 0.75 0.75 0 0 236

Oven 2.25 2.25 2.25 2.25 0 2.25 125

Total 1547

Page | 52

0 5 10 15 20 25200

400

600

800

1000

1200

1400

Loa

d (k

Wh)

Time (h)

Forecasted load Objective load Load after shifting

Figure 3. 2 Load curve for the residential area

0 5 10 15 20 250

5000

10000

15000

20000

25000

Cos

t ($)

Hours

Without DSM With DSM

Figure 3. 3 Hourly cost curve for the residential area

Page | 53

Table 3. 3 Simulation results of BPSO based load shifting [37]

Without DSM With DSM Reduction (%)

Residential area

PAR 1.85 1.38 25.41

Peak demand (kWh) 1363.6 1018 25.35

Total cost (US$) 2302.88 1914.4 16.87

Industrial area

PAR 1.62 1.53 5.56

Peak demand (kWh) 2727.3 2225 18.42

Total cost (US$) 5712.05 3923.48 31.31

Commercial area

PAR 1.7 1.31 22.91

Peak demand (kWh) 1818.2 1404 22.84

Total cost (US$) 3626.64 3022.3 16.66

3.3.3 Analysis of industrial appliances

A total of 133 devices of 6 types were considered for the simulation in the case of

industrial area as presented in Table 3.4. Industrial area has a lesser number of appliances

compared to residential and commercial areas. As can be seen from Figure 3.4, the load curve

after shifting closes to the objective curve except for the first 8 hours. At the 11th period, the

peak demand was 2727.3 kWh without DSM and reduced to 727.3 kWh with DSM. In the

case of the DSM strategy, the peak demand was 2225 kWh and shifted to 9th hour. Figure 3.5

shows the hourly cost reduction for the industrial area. In the peak periods, the hourly cost

with DSM significantly reduced with slight fluctuations. The results of the simulation suggest

that the proposed DSM scheme reduces the peak demand, PAR, and electricity cost by

shifting the load from peak hours to off-peak hours. In this case, the reduction of peak load,

PAR, and the total cost was 18.42%, 5.56%, and 31.31%, respectively.

Page | 54

Table 3. 4 Data of industrial area devices [37]

Load types Power consumption of load (kWh) Number of

devices 1st

hour

2nd

hour

3rd

hour

4th

hour

5th

hour

6th

hour

7th

hour

8th

hour

Heat pump

heat strips

10 10 10 0 0 0 0 0 14

Electric

furnace

10.5 10.5 25 10.5 10.5 0 0 0 20

Heat pump 0 0 0 9.77 9.77 9.77 9.77 9.77 18

Central AC 3 0 3 0 3 0 3 0 4

Electric

water heater

500 500 500 500 0 0 0 0 45

Freezer 32 32 32 32 32 0 32 32 32

Total 133

0 5 10 15 20 25500

1000

1500

2000

2500

3000

Loa

d (k

Wh)

Time (h)


Figure 3. 4 Load curve for the industrial area

Page | 55

0 5 10 15 20 250

10000

20000

30000

40000

50000

60000

70000

80000C

ost (

$)

Hours


Figure 3. 5 Hourly cost curve for the industrial area

3.3.4 Analysis of commercial appliances

Simulation was carried out in a total of 702 devices of 9 types in the case of

commercial area. The commercial appliances have a higher consumption and long period of

operation compared to industrial and residential areas. The number of the controllable device

is higher than the industrial area but lower than the residential area. As can be observed from

Figure 3.6, the shifted load curve is close to the objective with a slight variation. This

variation can be seen at 1st 6 hour, 12th hour to 16th hour and 23rd hour to 24th hour.

Starting time and ending time of operation of the devices is the reason behind this variation.

Without DSM, the peak demand was 1818.2 kWh at 18th hour. After DSM implementation,

the peak demand shifted to 17th hour and the value was 1404 kWh. The amount of peak load

reduction was 414.2 kWh and the percentage of peak load reduction was 22.84%. It can be

seen from Figure 3.6 that the electricity consumption is higher where the electricity price is

Page | 56

higher. Without DSM at 11th hour to 13th hour, the average electricity price was 24.9 ct/kWh

and the total average electricity consumption was 1558.4 kWh. The total average electricity

cost in that period was 388.87 USD. After DSM implementation, the total average electricity

consumption at 11th hour to 13th hour reduced to 705.367 kWh and the total average

electricity cost reduced to 180.11 USD. So the average electricity cost reduced in 11th hour

to 13th hour by 53.68%. Without DSM, the PAR and total electricity cost was 1.7 and

3626.64 USD, respectively. With the implementation of DSM, the PAR and total electricity

cost decreased to 1.31 and 3022.3 USD, respectively. Therefore, our proposed load shifting

technique able to reduce the PAR and total cost by 22.91% and 16.66%, respectively. Figure

3.7 shows the hourly cost curve for the commercial area. In the peak periods, the hourly cost

significantly reduced after DSM implementation. However, hourly electricity cost increased

in 1st hour to 9th hour due to the shifting of loads from peak hours during these hours.

Table 3. 5 Data of commercial area devices [37]

Load types Power consumption of loads in kWh Number of

devices 1st hour

2nd hour

3rd hour

4th hour

5th hour

6th hour

7th hour

8th hour

9th hour

10th hour

Broiler 88 88 0 88 88 88 0 0 88 0 89 Dish washer 32 32 32 0 32 32 0 0 32 0 110 Roster 55 55 55 0 55 55 0 0 55 0 93 Oven (Self cleaner)

65 65 65 65 65 65 0 0 65 0 78

Coffee maker 2 2 2 2 2 2 2 2 2 2 60 Hot plate 4 4 4 4 4 4 4 4 4 4 13 Oven 26 26 26 0 26 26 0 0 26 0 128 Bottle warmer 0 7 7 0 7 7 0 0 7 0 52 Trash compactor

29 29 29 0 29 29 0 0 29 29 79

Total 702

Page | 57

0 5 10 15 20 25200

400

600

800

1000

1200

1400

1600

1800

2000

Loa

d (k

Wh)

Time (h)


Figure 3. 6 Load curve for the commercial area

0 5 10 15 20 250

10000

20000

30000

40000

50000

Cos

t ($)

Hours


Figure 3. 7 Hourly cost curve for the commercial area

Page | 58

3.3.5 Comparative analysis with GA-DSM

Figure 3.8 shows the comparison between the proposed BPSO based and GA based

DSM method. GA based DSM method used the “flexible load shape” technique while BPSO

method used the “load shifting” technique. It can be observed from Fig. 5 that the GA based

DSM method reduced peak demand by 23.81%, 17.49%, and 19.29% for residential,

industrial, and commercial areas, respectively [86]. On the other hand, the proposed BPSO

based DSM method reduced the peak demand by 25.35%, 18.42%, and 22.84% for

residential, industrial, and commercial areas, respectively. Therefore, the BPSO based DSM

method improves the performance by 1.54%, 0.93%, and 3.55% for residential, industrial,

and commercial areas, respectively. In addition, the proposed BPSO based load shifting

method is simple in mathematical formulation and provides the maximum efficiency and

successive rate compared to GA based approach. However, BPSO is associated with some

limitations for instances, difficulties in chossing appropriate parameters for optimization,

sometimes it gets stuck in local optimum solution.

Residential Industrial Commercial0

5

10

15

20

25

30

Peak

load

red

eact

ion

(%)

BPSO GA

Figure 3. 8 Comparison between BPSO and GA based DSM

Page | 59

3.4 Conclusions

In this work, a BPSO based load shifting strategy is proposed, which has the potential

to bring the benefit for both customers and suppliers. This method is able to provide a stable

cost reduction curve. Simulation is carried out with a number of appliances of various types

in the residential, commercial, and industrial areas. It has been shown that the proposed

BPSO based method finds an optimal load schedule in terms of reduction of peak demand,

electricity price, and PAR. In addition, a comparative study has been carried out with the GA

based DSM method and found that the proposed BPSO based DSM method performs better.

Page | 60

CHAPTER 4: OPTIMAL MANAGEMENT OF HOME LOADS WITH RENEWABLE ENERGY INTEGRATION AND DEMAND RESPONSE STRATEGY

Abstract: The implementation of proper energy management techniques and utilization of

renewable energy resources enhance the energy efficiency and stability of future grid

systems. This research proposed a home energy management model consisting of microgrid

framework and demand side management (DSM) technique. To reduce peak load, peak to

average, and energy cost, households’ loads were shifted on the basis of price-based tariff

such as flexible and time of use tariff. Simulation was carried out using binary particle swarm

optimization algorithm in MATLAB. The microgrid was mathematically modelled, and the

impacts of DSM integrated microgrid were analysed for different households in terms of

electricity cost reduction. Simulations suggested that DSM implementation significantly

reduced peak loads and renewable resources produced significant trade-off. The proposed

integrated approach reduced 90%–100% of the total electricity cost of households.

Keywords: Demand side management; Demand response; Load shifting technique; BPSO

algorithm; Energy tariff; Renewable energy generation

Page | 61

4.1 Introduction

The increase in energy consumption as well as the rapid growth of population and the

lack of implementation of proper management techniques result in an extreme spike in

energy demand [3,167]. Existing electrical grid worsens the situation due to their old-

fashioned design as well as redundant and overstressed infrastructure [168]. Recently

environmental pollution gains much attention among the scientists and environmentalists

because of public consciousness of reducing carbon emission and political pressure [169].

About 85% of the total global energy consumption depends on fossil fuels [170]. The

excessive usage of fossil fuels is associated with the release of substantial CO2 emission. The

integration of RES in power generation is the most effective and feasible way to promote

sustainable development and reduce environmental pollutions [171].

The utilization and optimization of RES lead to the concept of microgrid as a

replacement for fossil energy sources [172]. Recently, the use of microgrid system has gained

significant popularity in finding ways to increase the stability of energy supply by integrating

distributed energy resources, such as wind turbines and solar panels, and distributed energy

storage like batteries [173]. Moreover, microgrid has distinct features, including reliability,

low investment costs, and regulations of different distributed generator units’ output voltage

and current [174]. Additionally, microgrid can be incorporated with different DSM

frameworks and operated in grid connected and off-grid modes.

DSM refers to the amendments of consumers’ energy consumption pattern to enhance

the efficiency of electrical energy systems and network [175,176]. This technique modifies

daily energy consumption pattern to achieve a desired load profile [177]. Numerous scholars

studied the implementation of DSM in residential loads management [178]. For instance,

Gottwalt et al. carried out a simulation for shifting residential loads on the basis of TOU tariff

[179]. Ma et al., Ozkan, Steen et al., Lu et al., Missaoui et al., and Ogunjuyigbe et al. also

Page | 62

developed DSM model to reduce peak load and energy cost [180–185]. Bharathi et al. solved

DSM optimization problems for residential loads management using genetic algorithm to

reduce peak load [86]. In our previous work, we analyzed the load shifting problem for

residential loads using the BPSO algorithm [186].

Modeling of microgrid with the application of DSM can play a pivotal role in

reducing peak load, energy inefficiency, and operational cost of electricity provider, thus

helping to reduce carbon foot prints. This integrated approach can reduce the amount of

energy required to buy from the grid. Recent works were dedicated to the implementation of

DSM along with the integration of renewable energy for achieving a balance between energy

generation and consumption. Quiggin et al. implemented the demand response program in a

residential microgrid with the integration of solar photovoltaic, wind turbine, and energy

storage [144]. This method optimized the energy balance between supply and demand by

reducing peak demand fluctuations by 16%, thus significantly reducing CO2. The effect of

DSM technique was also investigated in an industrial microgrid by Blake and O'Sullivan

[151], who proposed a method that reduced the overall electricity cost by 73% with almost

88% CO2 reduction. Palma-Behnke et al. modelled a microgrid system with solar

photovoltaic, wind turbine, diesel generator, battery bank, and water supply system [152].

The DSM mechanism along with artificial neural network was used for determining optimal

operation and reducing costs . Shen et al. carried out an incentive-based demand response

program in a microgrid, which consisted of micro turbines, wind turbine, fuel cells, solar

photovoltaic, storage devices, and controllable load [187]. The authors solved an operational

scheduling problem to achieve an optimal scheduling of the batteries and diesel generators.

Nunna and Doolla proposed an intelligent energy management system with demand response

program to reduce peak load demand in the microgrid, and they performed simulation by

using Java Agent Development framework [188]. To reduce electricity cost, an incentive

Page | 63

method was recommended by the authors to motivate the consumers for participating in the

demand response program. Philippou et al. investigated a price-based DSM mechanism and

performed sensitivity analysis to reduce peak load during summer and winter [157].

A comprehensive case study was carried out in the current work, where a price-based

(i.e., flexible and time of use tariff) demand response was applied for households’ load

scheduling. The DSM optimization problem was solved using BPSO algorithm in MATLAB.

Additionally, the work integrated renewable energy resources with DSM to reduce electricity

cost and maximize renewable energy use.

4.2 Load modeling and DSM implementation for multiobjective optimization

DSM implementation aims to reduce load demand during peak hours, reduce

electricity bill, and maximize the use of renewable energy as well as reduce the usage of

electricity from the main distribution grid. A BPSO-based load scheduling mechanism was

applied to manage residential load demand. In BPSO algorithm, each hour in a day was

denoted by row vector. The proposed DSM technique scheduled households’ shiftable

appliances. Each of the household devices considered in this study consumed different

amounts of energy with various power ratings. In the BPSO algorithm, a set of household

devices is presented by D= {d1, d2,……,dn}.

For each household, the scheduling vector of energy consumption of the appliances

can be presented by 𝐷𝑛=[𝑑11, ……… , 𝑑𝑛𝑡 ], where 𝑑𝑛1 is the energy consumption of appliances

n, scheduled for 1 hour. The total energy consumption for the household is estimated by the

following expression:

𝐸𝑛 =∑𝑑𝑛𝑡

24

𝑡=1

(4.1)

Page | 64

The devices were scheduled based on their daily energy consumption. The scheduled

energy consumption of the appliances can be expressed as follows:

𝐸𝑛 = ∑ 𝑑𝑛𝑡

𝐸𝑇

𝑡=𝑆𝑇

(4.2)

where ST and ET are the schedule start and end time of the appliances, respectively.

The cost function 𝐶𝑇𝑂𝑈, which represents the cost of energy in each hour, was

calculated based on flexible pricing tariff and time of use (TOU) tariff provided by the utility

company. Flexible pricing tariff introduced three periods, such as peak, off-peak, and

shoulder while, TOU tariff considered only peak and off-peak periods. Table 1 shows the

typical TOU tariff and flexible pricing tariff of Victoria, Australia

[https://www.canstarblue.com.au/electricity/victoria-electricity-tariffs/]. The price of

electricity at peak hour is higher as compared to price at off-peak hour due to the high

demand at peak hour. The peak and off-peak hours for TOU tariff are the same for weekdays

and weekends. In the case of flexible tariff, no peak period for weekends exists. However,

shoulder periods are extended from 7am to 10pm.

Table 4. 1 Typical electricity tariff (average value) in Victoria, Australia

Flexible TOU

Peak Off-peak Shoulder Peak Off-peak

Time 3pm–9pm 10pm–7am 9pm–10pm and 7am–3pm 7am–11pm 11pm–7am

Rate (ct/kWh) 45 19.5 35.5 36.5 17

The time span of 24 hours was divided into equal time slots, where t ∈ 𝑇. The cost is

the function of amount of energy consumed. The cost function (CTOU) for TOU tariff is

presented by the following equation:

https://www.canstarblue.com.au/electricity/victoria-electricity-tariffs/

Page | 65

𝐶𝑇𝑂𝑈 =

{

∑ 𝐸𝑛

𝑇 ∗ 𝑀𝑇𝑂, 𝑜𝑓𝑓 − 𝑝𝑒𝑎𝑘 ℎ𝑜𝑢𝑟

𝑇𝑜

𝑇=23

∑𝐸𝑛𝑇 ∗ 𝑀𝑇𝑃,

𝑇𝑝

𝑇=7

𝑝𝑒𝑎𝑘 ℎ𝑜𝑢𝑟𝑠

(4.3)

where MTP and MTO are the costs of electricity per unit for TOU tariff during the peak

and off-peak periods, respectively. In the case of TOU tariff, the total cost (CTT) for 24 hours

duration is the summation of cost during peak and off-peak hours, as shown in the following

equation:

𝐶𝑇𝑇 = 𝐶𝑇𝑃 + 𝐶𝑇𝑂 (4.4)

where CTP and CTO are the energy consumption cost during peak and off-peak hours,

respectively. In the case of flexible pricing tariff, the cost function (CFLP) is given as follows.

𝐶𝐹𝐿𝑃 =

{

∑ 𝐸𝑛

𝑇 ∗ 𝑀𝐹𝑂, 𝑜𝑓𝑓 − 𝑝𝑒𝑎𝑘 𝑝𝑒𝑟𝑖𝑜𝑑

𝑇𝑜

𝑇=22

∑ 𝐸𝑛𝑇 ∗ 𝑀𝐹𝑆, 𝑆ℎ𝑜𝑢𝑙𝑑𝑒𝑟 𝑝𝑒𝑟𝑖𝑜𝑑

𝑇𝑠

𝑇=21,7

∑ 𝐸𝑛𝑇 ∗ 𝑀𝐹𝑃 , 𝑝𝑒𝑎𝑘 𝑝𝑒𝑟𝑖𝑜𝑑

𝑇𝑝

𝑇=15

(4.5)

where MFP, MFS, and MFO are the per unit electricity price for flexible pricing tariff

during peak, shoulder, and off-peak hours, respectively. For flexible pricing tariff, the total

cost of energy consumption (CFT) is the summation of cost during peak, shoulder, and off-

peak hours and expressed as follows:

𝐶𝐹𝑇 = 𝐶𝐹𝑃 + 𝐶𝐹𝑆 + 𝐶𝐹𝑂 (4.6)

The load scheduling was formulated as a minimization problem, and the optimal

scheduling of the household appliances was obtained by solving the minimization problem as

Page | 66

follows, where, CT represents the total electricity cost, which can be either based on the

flexible pricing tariff or TOU pricing tariff.

Minimize∑𝐶𝑇

𝑇

𝑡=1

In BPSO algorithm, the particles are initialized randomly with the binary value

between 0 and 1. The status of load defined by 𝑥𝑖 = [𝑥1,𝑥2, …… . . 𝑥𝑛], ∀𝑥𝑖𝜖 (0,1), where 𝑥𝑖 =

1 represents the connection of the load, and 𝑥𝑖 = 0 indicates the disconnection of the load.

The BPSO algorithm optimized the objective function (i.e., minimization of cost function) by

updating the position and the velocity of the particles. The velocity of the particle is

expressed by 𝑥𝑖(𝑡𝐵𝑃𝑆𝑂+1) = 𝑥𝑖

𝑡𝐵𝑃𝑆𝑂 + 𝑉𝑖(𝑡𝐵𝑃𝑆𝑂+1), and the velocity of the particle is presented

as follows:

𝑉𝑖(𝑡𝐵𝑃𝑆𝑂+1) = 𝑤. 𝑉𝑖

𝑡𝐵𝑃𝑆𝑂 + 𝐶1. 𝑟𝑎𝑛𝑑(). (𝑥𝑝𝑏𝑒𝑠𝑡.𝑖𝑡𝐵𝑃𝑆𝑂 − 𝑥𝑖

𝑡𝐵𝑃𝑆𝑂) + 𝐶2 . 𝑟𝑎𝑛𝑑(). (𝑥𝑔𝑏𝑒𝑠𝑡.𝑖𝑡𝐵𝑃𝑆𝑂 − 𝑥𝑖

𝑡𝐵𝑃𝑆𝑂) (4.7)

where C1 and C2 are the cognitive constant, and w is the weighting function. To

schedule the household appliances, the objective function is fed to the BPSO algorithm on

basis of electricity price.

4.3 Microgrid Modeling

The proposed microgrid consists of three subsystems, namely, energy generation,

residential load demand, and energy distribution subsystems. The schematic diagram of a

hybrid microgrid is shown in Figure 4.1. The WT and solar PV panels worked as renewable

energy generators. In our proposed model, residential load profile was used as a demand

subsystem. The microgrid was connected to the grid through AC bus, and whole microgrid

system worked as a power distribution subsystem. DSM technique was applied in the

Page | 67

residential loads to schedule the load demand. The energy generated from the renewable

sources was used to meet the scheduled load demand on an hourly basis.

Wind turbine

Solar PV

Power electronic interface

Residential loads

AC Bus

DC/AC

Main grid

Renewable energy sources

DSM implementation

Figure 4. 1 Conceptual design of proposed microgrid model

4.3.1 PV system

The PV system produces electrical energy from solar energy, and the output of the PV

system depends on solar irradiance, efficiency of the PV panel, and atmospheric temperature

[189]. The output power from solar PV was calculated by the following equation:

PVout = PVr ∗ 𝐹𝑃𝑉 (4.8)

where PVout is the output power of solar PV. PVr and FPV indicate the rated power of

solar PV (kW) and performance of the solar PV (%), respectively. The performance of the

solar PV data was based on the Victorian state performance on the particular date [https://pv-

map.apvi.org.au/live].

4.3.2 Wind turbine

The wind power system converts the wind speed into electrical power. The output

power of the WT energy system depends on the hourly wind speed [190]. In this study, the

hourly wind speed data for the study area was collected from timeanddate.com

https://pv-map.apvi.org.au/live

https://pv-map.apvi.org.au/live

Page | 68

[https://www.timeanddate.com/weather/australia/melbourne/historic?month=11&year=2019].

The output power from WT can be calculated by the following equation [190,191]:

𝑃𝑊𝑇𝑂𝑈𝑇 =

{

0, 𝐶 < 𝐶𝐼𝑎𝑛𝑑 𝐶 ≥ 𝐶𝑜

𝑃𝑊𝑇𝑅 ∗𝐶 − 𝐶𝐼𝐶𝑅 − 𝐶𝐼

, 𝐶𝐼 ≤ 𝐶 < 𝐶𝑅

𝑃𝑊𝑇𝑅 , 𝐶𝑅 ≤ 𝐶 < 𝐶0

(4.9)

where PWTOUT and PWTR indicate the output power from WT and rated power of

WT, respectively. C, CR, CI, and CO represent the real time wind speed, rated wind speed,

cut-in speed, and cut-out speed, respectively.

4.3.3 Energy savings from renewables

The hourly power generated from the solar PV and WT is compared with the

scheduled load demand for each of the households. The hourly surplus energy (i.e., excess

energy from renewables after fulfilling the demand) and energy deficit (i.e., energy required

to purchase from grid after consuming the energy from renewables) were estimated using the

following equation:

𝐸𝑆/𝐷 = 𝑃𝑉𝑂𝑈𝑇𝑡 + 𝑃𝑊𝑇𝑂𝑈𝑇

𝑡 − 𝐸𝑛𝑡 (4.10)

where ES/D is amount of energy surplus or deficit in an hour. The positive value of

ES/D indicates energy surplus, whereas negative value represents energy deficit. The monthly

revenue generation by selling surplus energy can be estimated using the following equation,

where FRAV is the average feed-in-tariff in Victoria:

𝑅𝑀 = [∑(∑(𝑃𝑉𝑂𝑈𝑇𝑡 + 𝑃𝑊𝑇𝑂𝑈𝑇

𝑡 − 𝐸𝑛𝑡)

24

𝑡=1

)

30

𝑑=1

] ∗ 𝐹𝑅𝐴𝑉; 𝑃𝑉𝑂𝑈𝑇𝑡 + 𝑃𝑊𝑇𝑂𝑈𝑇

𝑡 > 𝐸𝑛𝑡 (4.11)

https://www.timeanddate.com/weather/australia/melbourne/historic?month=11&year=2019

Page | 69

4.4 Results and discussion

The performance of the proposed microgrid model was investigated for a case study.

The simulation was carried out in four households in Victoria, Australia. The electricity

consumption of each of the households and different set of appliances are listed in Table 4.2.

In this mode, shiftable and nonshiftable loads are considered for the estimation of overall

energy requirement and cost. However, only shiftable loads were considered for load shifting

simulation. Apart form this, high-power appliances such as vacuum cleaner, washing

machine and iron was used in weekends. The electricity consumption of these devices were

included in the weekends’ load profile.

Table 4. 2 Appliances and power consumption pattern for households

Device Preferred time of use Hourly consumption (kW) Power rating (kW) 1 h 2 h 3 h

Household 1 Washing machine 10am–5pm 0.25 0.25 0 0.25 Vacuum cleaner 12pm–3pm 0.9 0 0 1.8 Oven 6am–9am and 12:30pm–

1:30pm 0.37 0.37 0 1.1

Iron 7pm–9pm 0.34 0 0 1 Rice cooker 5pm–7pm 0.25 0.25 0 0.5 Induction cooker 5pm–8pm 0.2 0.2 0.1 0.2 Room heater 6pm–11pm 0.95 0.95 0.95 1.9 Toaster 7am–9 am 0.17 0 0 0.7 Household 2 Washing machine 11am–2pm 1.25 0.625 0 1.25 Vacuum cleaner 10am–1pm 1 0 0 2 Iron 5pm–10pm 0.67 0 0 2 Rice cooker 6am–9am and 7pm–9pm 0.35 0.35 0 0.7 Dish washer 5pm–7pm 1.8 0 0 1.8 Room heater 7pm–10pm 2 1 2 2 Toaster 7am–9am 0.29 0 0 1.75 Hair dryer 7am–8am 0.36 0 0 2.2 Air Fryer 5pm–7pm 0.5 0 0 1.5 Oven 6am–9am 0.25 0.25 0.25 1.5 Household 3 Washing machine 9am–1pm 0.25 0.041 0 0.25 Vacuum cleaner 10am–12pm 1 0 0 2 Blender 6am–9am and 6pm–8pm 0.16 0 0 0.5 Iron 7pm–10pm 0.66 0 0 2

Page | 70

Rice cooker 7am–9am and 6pm–9pm 0.35 0.35 0 0.7 Room heater 6pm–10pm 1.9 1.9 0.9 1.9 Toaster 7am–9am 0.13 0.13 0 0.8 Oven 8am–9am 0.23 0.23 0.23 1.4 Household 4 Vacuum cleaner 9am–4pm 1.2 0 0 1.2 Blender 4pm–6pm 0.1 0.1 0 0.3 Iron 9pm–10pm 0.8 0 0 2.4 Rice-cooker 6pm–8pm 0.35 0.35 0 0.7 Double hot plate 6pm–9pm 0.7 0.7 0 1.4 Room heater 7pm–11pm 1.9 1.9 1.9 1.9 Oven 7am–10am 0.2 0.2 0.2 1.2

4.4.1 Load profiles and scheduling of the loads

Figure 4.2 shows the weekday’s average load profiles for four households and their

comparison with the load profiles after DSM implementation under the flexible pricing and

TOU tariff scheme. The weekdays load data was taken from one random weekday of each

week for the whole last year (i.e., in 2018). Then the load profile was developed using the

average of 52 weeks’ data for each household. The loads were shifted based on electricity

price and consumers’ preference keeping the total load demand same. Figure 4.2 shows that

the each of the households’ hourly electricity consumption was different because of variation

in preference and behaviour of energy consumption. Whether the consumers used the flexible

pricing tariff or TOU tariff scheme, they were not getting any benefit from the tariff scheme

because most of the electricity was consumed during peak periods. In the case of flexible

pricing tariff, the peak period was between 15th hour to 21th hour, whereas the peak period

was between 7th hour to 23rd hour for TOU pricing tariff. In the peak period, the electricity

providers charged a higher price for per unit of electricity compared to off-peak and shoulder

periods. In addition,, the total electricity demand in peak period was about 8.24 kWh for the

household 1 under flexible pricing tariff. After the implementation of DSM, the total energy

demand during the peak period was reduced to 6.41 kWh due to the shifting of loads from

peak to off-peak and shoulder hours. Therefore, almost 22.2% (1.83 kWh) of loads in peak

Page | 71

periods was shifted from peak periods to off-peak and shoulder periods, where the per unit

electricity price was lower than that during peak periods. Similarly, the implementation of

DSM shifted approximately 18.2%, 23.4%, and 20.6% of loads in peak periods from peak

periods to off-peak and shoulder periods for households 2, 3, and 4, respectively.

Additionally, the individual peak load for all the households was also reduced after DSM

implementation, as shown in Figure 4.2. In the case of TOU pricing tariff, the energy

providers offer only peak and off-peak hour pricing scheme. Most of the hours of the day

(7am–11pm) are considered peak period, where consumers usually prefer to use their loads.

Therefore, no significant load shifting was observed in the case of TOU pricing tariff due to

the users’ preference of load utilization, nature of loads, and extended hours of peak periods.

However, the implementation of DSM reduced the individual peak load and distributed the

reduced amount of load in the entire peak periods for all the households, as presented in

Figure 4.2.

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.0

0.5

1.0

1.5

2.0

Loa

d (k

Wh)

Time (h)

Without DSM Flexible price Time of use

Household 1

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

Loa

d (k

Wh)

Time (h)


Household 2

Page | 72

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75L

oad

(kW

h)

Time (h)


Household 3

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

Loa

d (k

Wh)

Time (h)


Household 4

Figure 4. 2 Average hourly load profile of households during weekday

Figure 4.3 displays the weekend’s average hourly electricity consumption for the

households. The weekends load data was taken from one random weekend of each week for

the whole last year (i.e., in 2018). Then the load profile was developed using the average of

52 weeks’ data for each household. The average electricity consumption in the weekend for

all the households was higher than in weekdays. In the case of flexible tariff, DSM was not

applied for weekend because of the absence of peak period for weekend. Moreover, a

shoulder rate is charged most of the day (7am–10pm). Similar to weekday, peak and off-peak

rates were charged in the weekend for TOU pricing tariff. Hence, DSM was applied to shift

the loads from peak periods. However, no notable load shifting was obtained for all the cases

because of the similar reasons, as described earlier. When DSM was applied, the individual

peak load was reduced significantly and distributed throughout peak periods. However, the

individual peak load remained in the same hour for all the households, as illustrated in Figure

4.3.

Page | 73

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.0

0.5

1.0

1.5

2.0

2.5

3.0L

oad

(kW

h)

Time (h)

Without DSM TOU

Household 1

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:00

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Loa

d (k

Wh)

Time (h)

Without DSM TOU

Household 2

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.0

0.5

1.0

1.5

2.0

2.5

3.0

Loa

d (k

Wh)

Time (h)

Without DSM TOU

Household 3

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000.0

0.5

1.0

1.5

2.0

2.5

3.0

Loa

d (k

Wh)

Time (h)

Without DSM TOU

Household 4

Figure 4. 3 Average hourly load profile of households during weekend

4.4.2 Performance gain in terms of energy and cost

Figure 4.4 depicts the effects of DSM implementation on the reduction of electricity

cost in weekday for all the households considered. The DSM allows the consumers to shift

the load from peak hours to off-peak and shoulder hours as discussed earlier, causing

substantial electricity cost (Figure 4.4). The reduction of hourly electricity cost depends on

the type of selected appliances and the flexibility of the operation time, which were provided

by the consumers for load scheduling. The hourly electricity cost presented in Figure 4.4

shows that the electricity cost in peak periods was about 65.1% of the total cost for household

1, which was reduced to approximately 50.7% after DSM application. Thus, the total

electricity consumption was significantly reduced during peak periods. In the case of

Page | 74

household 1, TOU and flexible pricing tariff showed higher electricity expense in the 20th

hour due to the peak demand in that hour. Similarly, the electricity cost in peak periods was

reduced from 59.7%, 52.7%, and 51.7% to 51.6%, 42.3%, and 43.2% of the total cost under

flexible pricing tariff for households 1, 2, and 3, respectively. In the case of TOU pricing

tariff, a slight reduction in electricity cost during peak periods was observed for all the

households.

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

20

40

60

80

100 Flexible price without DSM Flexible price with DSM TOU without DSM TOU with DSM

Ele

ctri

city

cos

t (ct

)

Time (h)

Household 1

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

16

32

48

64


Ele

ctri

city

cos

t (ct

)

Time (h)

Household 2

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

15

30

45

60


Ele

ctri

city

cos

t (ct

)

Time (h)

Household 3

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

15

30

45

60


Ele

ctri

city

cos

t (ct

)

Time (h)

Household 4

Figure 4. 4 Hourly cost curves for average load of households during weekday

Figure 4.5 shows the hourly electricity cost reduction during weekend after load

scheduling, particularly under TOU pricing tariff. In the case of flexible pricing tariff, no load

was scheduled. Figure 4.5 shows that the pattern of hourly electricity cost of unscheduled

loads under flexible pricing tariff was almost similar to that of unscheduled loads under TOU

Page | 75

pricing tariff. When TOU tariff rate is considered, most of the household appliances were

operated during the peak periods. The cost of electricity consumed in peak periods under

TOU tariff was about 94.9%, 94.3%, 97.9%, and 97.5% of total electricity cost for

households 1, 2, 3, and 4, respectively. However, with the application of DSM, the cost of

electricity in peak periods for the households was reduced to 92.1%, 93.5%, 95.9%, and

95.5% of total electricity cost, respectively.

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

20

40

60

80

100 Flexible price without DSM TOU without DSM TOU with DSM

Ele

ctri

city

cos

t (ct

)

Time (h)

Household 1

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

20

40

60

80


Ele

ctri

city

cos

t (ct

)

Time (h)

Household 2

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

20

40

60

80

100


Elec

tric

ity c

ost (

ct)

Time (h)

Household 3

3:00 6:00 9:00 12:00 15:00 18:00 21:00 24:000

20

40

60

80

100


Elec

tric

ity c

ost (

ct)

Time (h)

Household 4

Figure 4. 5 Hourly cost curves for average load of households during weekend

The user always wants to reduce their electricity cost and prefers to maintain a

balance of load with a low PAR. In the present case study, the application of DSM showed a

significant reduction in electricity cost and PAR. Table 4.3 shows the comparison between

the flexible pricing and TOU pricing tariff scheme for load scheduling in terms of peak load

Page | 76

reduction, total electricity cost reduction, and PAR reduction. In weekdays, the PAR and

electricity cost reduction were 12.8% to 27.1% and 5.4% to 6.6% under flexible pricing tariff,

whereas those for TOU tariff were 12.8% to 20.4% and 2.0% to 2.9%, respectively, due to

the application of DSM. In the weekend, the load scheduling under TOU tariff reduced the

PAR and electricity cost by 14.9% to 20.1% and 1.0% to 2.9%, respectively.

Table 4. 3 Summary of the load shifting results in percentages

Consumers Flexible price tariff (Weekday) TOU tariff (Weekday) TOU tariff (Weekend)

PAR PL Cost PAR PL Cost PAR PL Cost

Household 1 27.1 27.1 6.6 16.1 16.1 2.7 19.0 19.0 2.9

Household 2 15.9 15.9 6.3 20.4 20.4 2.3 20.1 20.1 1.0

Household 3 20.1 20.1 5.4 12.8 12.8 2.9 14.9 14.9 2.2

Household 4 12.8 12.8 5.9 14.9 14.9 2.0 16.0 16.0 2.1

PAR- Peak to average ratio; PL- Peak load

4.4.3 Renewable energy integration

We considered residential microgrid, which can be connected to the main grid to

import and export power for each of the households. The proposed microgrid model contains

a 3 kW solar panel and a 1.5 kW wind turbine. The hourly power output from solar PV was

estimated using the Victorian hourly performance data of solar PV. According to the

technical specifications of the 1.5 kW wind turbine widely used in Australia

[https://www.australianwindandsolar.com/aws-hc-wind-turbines], the rated, cut-in, and cut-

out speeds are 10.5 m/s, 2.7 m/s, and 12.1 m/s, respectively. Hence, the hourly wind power

output was predicted based on the hourly wind speed. Figure 4.6 presents the average hourly

predicted power generation from the renewable energy sources for weekday and weekend.

The hourly power output for weekday and weekend showed almost the similar trend. The

higher renewable output predicted during the mid-day for both weekday and weekend was

https://www.australianwindandsolar.com/aws-hc-wind-turbines

Page | 77

due to the higher PV output, which was linked to the higher intensity of solar irradiance

during those periods.

0 5 10 15 20 250.0

0.5

1.0

1.5

2.0

Pow

er o

utpu

t (kW

)

Time (h)

Wind turbine Solar PV Total renewable

(A)

0 5 10 15 20 250.0

0.5

1.0

1.5

2.0

2.5(B) Solar PV

Wind turbine Total renewable

Pow

er o

utpu

t (kW

)Time (h)

Figure 4. 6 Average power output from renewable energy sources (A) Weekday (B) Weekend

Figure 4.7 illustrates the average hourly energy surplus and deficit with the

integration of renewable generation for each of the households after load shifting. In the case

of weekday, energy deficit was observed during 6am–9am and 18pm–22pm for all the

households under flexible and TOU pricing tariff due to the use of more loads and less

generation from renewables. The higher surplus energy was obtained during the mid-day for

both tariff cases associated with the higher renewable energy generation and less energy

consumption in those periods, as shown in Figure 4.7 (A) and (B). On the contrary, the

energy deficit for weekend occurred during 9am–10am and 15pm–23pm. Similar to weekday,

weekend also showed maximum energy surplus during mid-day. Therefore, approximately

7.65–11.51 kW of surplus energy was obtained per day for the studied households by the

incorporation of DSM and renewable energy resources. Accordingly, the estimated monthly

surplus energy was in the range between 223 kW and 275 kW which can be sold into grid to

generate revenue.

Page | 78

0 5 10 15 20 25

-1.0

-0.5

0.0

0.5

1.0

1.5

Deficit energy

Pow

er (k

W)

Time (h)

Household 1 Household 2 Household 3 Household 4

Surplus energy

(A)

0 5 10 15 20 25

-1.0

-0.5

0.0

0.5

1.0

1.5

Deficit energy

Pow

er (k

W)

Time (h)


Surplus energy

(B)

0 5 10 15 20 25

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Deficit energy

Pow

er (k

W)

Time (h)


Surplus energy

(C)

0 5 10 15 20 25-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Deficit energy

Pow

er (k

W)

Time (h)


Surplus energy

(D)

Figure 4. 7 Hourly energy surplus and deficit for each of the households after load shifting

(A) flexible pricing weekday, (B) TOU weekday, (C) flexible pricing weekend and (D) TOU weekend

4.4.4 Trade off from DSM integrated with microgrid

Figure 4.8 shows the monthly electricity cost and benefits gained from the DSM

implementation and renewable integration. Each of the households generated revenue by

selling surplus energy, and the value was estimated based on the average Victorian feed-in-

tariff of 10 ct/kWh [https://www.canstarblue.com.au/electricity/victoria-electricity-tariffs/].

Figure 4.8 (A) shows that the monthly electricity cost was reduced by 5.5%–6.1% for flexible

pricing tariff when some of the total household’s loads were shifted from peak hours to off-

peak and shoulder hours based on the electricity price and consumers’ preference. On the

contrary, the maximum electricity cost reduction in the case of TOU pricing tariff was only

just above 2.5%, as shown in Figure 4.8 (B). In the case of TOU pricing tariff, an

https://www.canstarblue.com.au/electricity/victoria-electricity-tariffs/

Page | 79

insignificant amount of load was shifted from peak hours to off-peak hours due to users’

preference and nature of loads, which were linked to a small reduction in electricity cost.

Analysis of flexible and TOU pricing tariff shows that users’ monthly electricity cost was

reduced significantly for both tariff cases (Figure 4.8). For instance, over 88% of base

electricity cost (i.e., electricity cost without DSM) reduction was obtained for household 4

under flexible pricing tariff due to the combined effects of load shifting and renewable

integration. Additionally, the monthly revenue generation from surplus energy was more than

15% of the electricity cost without DSM, which can compensate the users’ remaining

electricity cost after DSM implementation and renewable integration. Therefore, considering

all the scenarios i.e., load shifting, power generation from renewables and revenue from

surplus energy, the households considered for the current case study the need to spend from

no to only 9.5% of the total electricity cost without DSM.


20

40

60

80

100

120

140

160

180 Cost without DSM Cost with DSM Cost with DSM and renewables Revenue from renewables

Cur

renc

y (A

U$)

(A)


20

40

60

80

100

120

140

160

180 Cost without DSM Cost with DSM Cost with DSM and renewables Revenue from renewables

Cur

renc

y (A

U$)

(B)

Figure 4. 8 Monthly cost analysis under different scenarios (A) flexible pricing tariff (B) TOU pricing tariff

4.5 Conclusions

The current work aims to minimize the monetary expenses of electricity consumption

in households by shifting the loads to the times with lower electricity price and utilizing

energy from renewables. A residential household load management model was proposed

based on the price-based demand response along with the integration of renewable energy

Page | 80

resources. Simulation and case studies were conducted in different households to analyze the

effectiveness of the proposed model (i.e., DSM integration with renewables) under flexible

and TOU pricing schemes using BPSO algorithm in MATLAB. Results showed the potential

benefits of DSM implementation in reducing peak load and electricity cost. As a result of

demand response program and renewable energy integration, the consumers used a minimal

amount of electricity from grid, and they could sell surplus energy to the grid. This practice

greatly impacts the reduction of the households’ monthly electricity cost.

Page | 81

CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS

Conclusions:

This thesis focuses on the management of household load demand in the context of

SG and microgrid. In order to achieve an optimal schedule, load shifting and DR program

were implemented along with renewable energy integration based on real time data. The

scheduling problem was optimized in terms of reduction of peak load, electricity bill, and

PAR under various pricing schemes. BPSO algorithm was used for optimizing the proposed

load scheduling model in MATLAB. Microgrid was modelled mathematically for residential

households. The simulation was carried out to validate the proposed model under the

Victorian tariff rate. The following specific conclusions can be drawn from the simulation

results.

The BPSO based method finds an optimal load schedule in terms of reduction of peak

demand, electricity cost, and PAR for residential, industrial, and commercial loads.

The proposed BPSO based DSM method performs better over GA based approach in

terms of optimization for residential, industrial, and commercial loads scheduling.

DSM implementation provides potential benefits in terms of reduction of peak load

and electricity cost for residential households.

As a result of both demand response program and renewable energy integration, the

consumers used a minimal amount of electricity from the grid and they could sell

surplus energy to the grid.

Future work:

In this thesis, various problems of energy management have been investigated.

Various approaches have been suggested to provide a solution by attaining an optimal load

schedule in terms of cost and peak load reduction. But still, there are some prospective

Page | 82

directions to extend the current research work. Future research work can be extended as

follows:

In order to reduce the operational cost and emission of a smart grid, a probabilistic

model can be further investigated. Wind and solar energy are stochastic in nature and

always uncertain to predict the accurate generation. Due to this issue, a probability

distribution function can be introduced in microgrid modeling to predict the behavior

of renewable energy. Incentive based DR program could be another possible solution

to remove the uncertainties in SG.

Further extensive study can be performed considering a large number of smart homes

along with energy storage systems and large renewable energy system.

Adding electric vehicles to the home energy management system as a big load is

realistic nowadays, which regulates voltage and frequency. However, more research is

required on various aspects particularly, on the improvement of battery lifetime as

battery requires to undergo frequent charging and discharging, which causes

wearing.The customers can be classified into different subgroups for analyzing the

potential countermeasure of peak loads based on the various pricing schemes.

Hybrid algorithms can be developed and implemented for solving the DSM

optimization problems and establishment of comparison with existing algorithms.

Such investigations would be extremely useful to make the load management system more

additive and to achieve the desired goal.

Page | 83

References

[1] Halder PK, Parvez MS. Financial analyses and social impacts of solar home systems in

Bangladesh: A case study. Int J Renew Energy Res 2015;5:398–403.

[2] Khan I, Halder PK. Electrical energy conservation through human behavior change:

Perspective in Bangladesh. Int J Renew Energy Res 2016;6:43–52.

[3] Halder PK, Paul N, Joardder MUH, Sarker M. Energy scarcity and potential of

renewable energy in Bangladesh. Renew Sustain Energy Rev 2015;51:1636–49.

doi:10.1016/j.rser.2015.07.069.

[4] Warren P. A review of demand-side management policy in the UK. Renew Sustain

Energy Rev 2014;29:941–51. doi:10.1016/j.rser.2013.09.009.

[5] Sun Q, Ge X, Liu L, Xu X, Zhang Y, Niu R, et al. Review of Smart Grid

comprehensive assessment systems. Energy Procedia 2011;12:219–29.

doi:10.1016/j.egypro.2011.10.031.

[6] Mazlumi K, Gholiha MM, Abbasi R, Mazlumi F. Presenting a New Method Based on

Branch Placement for Optimal Placement of Phasor Measurement Units. J Oper

Autom Power Eng 2014;2:32–9.

[7] Li Q, Zhou M. The future-oriented grid-smart grid. J Comput 2011;6:98–105.

doi:10.4304/jcp.6.1.98-105.

[8] Agrawal P. Overview of DOE Microgrid Activities. Symp. Microgrids, Montreal:

2006, p. 1–32.

[9] Alizadeh M, Li X, Wang Z, Scaglione A, Melton R. Demand-side management in the

smart grid: Information processing for the power switch. IEEE Signal Process Mag

2012;29:55–67. doi:10.1109/MSP.2012.2192951.

[10] Gellings CW. The smart grid : enabling energy efficiency and demand response.

Fairmont Press; 2009.

[11] Metke AR, Ekl RL. Security technology for smart grid networks. IEEE Trans Smart

Grid 2010;1:99–107. doi:10.1109/TSG.2010.2046347.

[12] Tuballa ML, Abundo ML. A review of the development of Smart Grid technologies.

Renew Sustain Energy Rev 2016;59:710–25. doi:10.1016/j.rser.2016.01.011.

Page | 84

[13] Maharjan S, Zhang Y, Gjessing S, Tsang D. User-Centric Demand Response

Management in the Smart Grid with Multiple Providers. IEEE Trans Emerg Top

Comput 2014;6750:1–1. doi:10.1109/TETC.2014.2335541.

[14] Gellings CW. Evolving practice of demand-side management. J Mod Power Syst

Clean Energy 2017;5:1–9. doi:10.1007/s40565-016-0252-1.

[15] Qureshi WA, Nair N-KC, Farid MM. Impact of energy storage in buildings on

electricity demand side management. Energy Convers Manag 2011;52:2110–20.

doi:10.1016/J.ENCONMAN.2010.12.008.

[16] Wu Y, Ravey A, Chrenko D, Miraoui A. Demand side energy management of EV

charging stations by approximate dynamic programming. Energy Convers Manag

2019;196:878–90. doi:10.1016/J.ENCONMAN.2019.06.058.

[17] Wang Y, Yang Z, Mourshed M, Guo Y, Niu Q, Zhu X. Demand side management of

plug-in electric vehicles and coordinated unit commitment: A novel parallel

competitive swarm optimization method. Energy Convers Manag 2019;196:935–49.

doi:10.1016/J.ENCONMAN.2019.06.012.

[18] Awais M, Javaid N, Shaheen N, Iqbal Z, Rehman G, Muhammad K, et al. An efficient

genetic algorithm based demand side management scheme for smart grid. 18th Int.

Conf. Network-Based Inf. Syst., 2015, p. 351–6. doi:10.1109/NBiS.2015.54.

[19] Rahim S, Javaid N, Ahmad A, Khan SA, Khan ZA, Alrajeh N, et al. Exploiting

heuristic algorithms to efficiently utilize energy management controllers with

renewable energy sources. Energy Build 2016;129:452–70.

doi:10.1016/j.enbuild.2016.08.008.

[20] Zhao Z, Lee WC, Shin Y, Member S, Song K. An Optimal Power Scheduling Method

for Demand Response in Home Energy Management System. IEEE Trans Smart Grid

2013;4:1391–400.

[21] Khan MA, Javaid N, Mahmood A, Khan ZA. A generic demand-side management

model for smart grid. Int J Energy Res 2015;39:954–64. doi:10.1002/er.

[22] Arabali A, Ghofrani M, Fadali MS, Baghzouz Y. Genetic-Algorithm-Based

Optimization Approach for Energy Management. IEEE Trans Power Deliv

2013;28:162–70.

Page | 85

[23] Hussain H, Javaid N, Iqbal S, Hasan Q, Aurangzeb K, Alhussein M. An Efficient

Demand Side Management System with a New Optimized Home Energy Management

Controller in Smart Grid. Energies 2018;11:190. doi:10.3390/en11010190.

[24] Fang X, Misra S, Xue G, Yang D. Smart Grid — The New and Improved Power Grid: A Survey. IEEE

Commun Surv Tutorials 2012;14:944–80. doi:10.1109/SURV.2011.101911.00087.

[25] Farhangi H. The path of the smart grid. IEEE Power Energy Mag 2010;8:18–28.

doi:10.1109/MPE.2009.934876.

[26] Zhong J, Zheng R, Yang W, Wu F. Construction of Smart Grid at Information Age.

Power Syst Technol 2009;33:12–8.

[27] Yi T, Pipattanasomporn, Manisa Sheng-nan S, Hao-ming L, Rahman S. Comparative

study on smart grid related R&D in China, the United States and the European Union.

Power Syst Technol 2009;33:7–15.

[28] Rahman S, Rinaldy. An efficient load model for analyzing demand side management

impacts. IEEE Trans Power Syst 1993;8:1219–26. doi:10.1109/59.260874.

[29] Cohen AI, Wang CC. An optimization method for load management scheduling. IEEE

Trans Power Syst 1988;3:612–8.

[30] Macedo M, Galo J, Almeida L, Lima A. Opportunities and Challenges of DSM in

Smart Grid Environment. Third Int. Conf. Smart Grids, Green Commun. IT Energy-

aware Technol., 2013, p. 156–60.

[31] Nguyen H, Song J, Han Z. Demand side management to reduce peak-to-average ratio

using game theory in smart grid. IEEE Conf. Comput. Commun. Work., 2012, p. 91–6.

doi:10.1109/INFCOMW.2012.6193526.

[32] Zhang N, Hu Z, Dai D, Dang S, Yao M, Zhou Y. Unit Commitment Model in Smart

Grid Environment Considering Carbon Emissions Trading. IEEE Trans Smart Grid

2015;7:420–7. doi:10.1109/TSG.2015.2401337.

[33] Xin A, Xiao L, Wen Jie Q, Yang W. Bid-scheduling of demand side reserve based on

demand response considering carbon emission trading in smart grid. 5th Int. Conf.

Crit. Infrastruct., 2010, p. 1–6. doi:10.1109/CRIS.2010.5617534.

Page | 86

[34] Li D, Sun H, Chiu W-Y. Achieving Low Carbon Emission Using Smart Grid

Technologies. IEEE 85th Veh. Technol. Conf., 2017, p. 1–5.

doi:10.1109/VTCSpring.2017.8108624.

[35] Gellings CW, Chamberlin JH. Demand-side management: Concepts and methods.

1987.

[36] Yang H, Zhang Y, Tong X. System Dynamics Model for Demand Side Management.

3rd Int. Conf. Electr. Electron. Eng., 2006, p. 1–4. doi:10.1109/ICEEE.2006.251854.

[37] Logenthiran T, Srinivasan D, Shun TZ. Demand side management in smart grid using

heuristic optimization. IEEE Trans Smart Grid 2012;3:1244–52.

doi:10.1109/TSG.2012.2195686.

[38] Kinhekar N, Padhy NP, Gupta HO. Multiobjective demand side management solutions

for utilities with peak demand deficit. Int J Electr Power Energy Syst 2014;55:612–9.

doi:10.1016/j.ijepes.2013.10.011.

[39] Gellings CW. The concept of demand-side management for electric utilities. Proc

IEEE 1985;73:1468–70. doi:10.1109/PROC.1985.13318.

[40] Sinha A, De M. Load Shifting Technique for Reduction of Peak Generation Capacity

Requirement in Smart Grid. IEEE Int. Conf. Power Electron. Intell. Control Energy

Syst., 2016, p. 1–5.

[41] Gelazanskas L, Gamage KAA. Demand side management in smart grid: A review and

proposals for future direction. Sustain Cities Soc 2014;11:22–30.

doi:10.1016/j.scs.2013.11.001.

[42] Song L, Xiao Y, Van Der Schaar M. Demand side management in smart grids using a

repeated game framework. IEEE J Sel Areas Commun 2014;32:1412–24.

doi:10.1109/JSAC.2014.2332119.

[43] Ruiz N, Cobelo I, Oyarzabal J. A Direct Load Control Model for Virtual Power Plant

Management. IEEE Trans Power Syst 2009;24:959–66.

doi:10.1109/TPWRS.2009.2016607.

[44] Yao L, Member S, Chang W, Yen R. An Iterative Deepening Genetic Algorithm for

Scheduling of Direct Load Control. IEEE Trans Power Syst 2005;20:1414–21.

Page | 87

[45] Kothari DP, Nagrath IJ. Modern power system analysis. McGraw-Hill Higher

Education; 2008.

[46] Pina A, Silva C, Ferrão P. The impact of demand side management strategies in the

penetration of renewable electricity. Energy 2012;41:128–37.

doi:10.1016/j.energy.2011.06.013.

[47] Infield DG, Member S, Short J, Horne C, Freris LL. Potential for Domestic Dynamic

Demand-Side Management in the UK. IEEE Power Eng. Soc. Gen. Meet., 2007, p. 1–

6.

[48] Bayod-Rújula AA. Future development of the electricity systems with distributed

generation. Energy 2009;34:377–83. doi:10.1016/j.energy.2008.12.008.

[49] Torriti J, Hassan MG, Leach M. Demand response experience in Europe: Policies,

programmes and implementation. Energy 2010;35:1575–83.

doi:10.1016/j.energy.2009.05.021.

[50] Wang J, Bloyd CN, Hu Z, Tan Z. Demand response in China. Energy 2010;35:1592–7.

doi:10.1016/j.energy.2009.06.020.

[51] Chu W-C, Chen B-K, Fu C-K. Scheduling of Direct Load Control to MiniDice for A

Utility Suffering from Generation Load Reduction Shortage. IEEE Trans Power Syst

1999;8:1525–30. doi:10.1109/59.260955.

[52] Sheblt GB. Direct Load Control - a Profit-Based Load Management Using Linear

Programming. IEEE Trans Power Syst 1998;13:688–95. doi:10.1109/59.667401.

[53] Conejo AJ, Morales JM, Baringo L. Real-time demand response model. IEEE Trans

Smart Grid 2010;1:236–42. doi:10.1109/TSG.2010.2078843.

[54] Pipattanasomporn M, Kuzlu M, Rahman S, Member S, Kuzlu M, Rahman S. An

Algorithm for Intelligent Home Energy Management and Demand Response Analysis.

IEEE Trans Smart Grid 2012;3:1–8. doi:10.1109/TSG.2012.2201182.

[55] Logenthiran T, Srinivasan D, Vanessa KWM. Demand side management of smart grid:

Load shifting and incentives. J Renew Sustain Energy 2014;6. doi:10.1063/1.4885106.

[56] Siano P. Demand response and smart grids - A survey. Renew Sustain Energy Rev

2014;30:461–78. doi:10.1016/j.rser.2013.10.022.

Page | 88

[57] Yang P, Chavali P, Gilboa E, Nehorai A. Parallel load schedule optimization with

renewable distributed generators in smart grids. IEEE Trans Smart Grid 2013;4:1431–

41. doi:10.1109/TSG.2013.2264728.

[58] Qdr Q. Benefits of demand response in electricity markets and recommendations for

achieving them. Washington: 2006.

[59] Wu Q, Wang L, Cheng H. Research of TOU Power Price Based on Multi-Objective

Optimization of DSM and Costs of Power Consumers. IEEE Int. Conf. Electr. Util.

Deregulation, Restruct. Power Technol., 2004, p. 343–8.

[60] Ortega-vazquez MA, Kirschen DS. Estimating the Spinning Reserve Requirements in

Systems With Significant Wind Power Generation Penetration. IEEE Trans Power

Syst 2009;24:114–24.

[61] Palensky P, Dietrich D. Demand side management: Demand response, intelligent

energy systems, and smart loads. IEEE Trans Ind Informatics 2011;7:381–8.

doi:10.1109/TII.2011.2158841.

[62] Baharlouei Z, Hashemi M. Demand Side Management Challenges in Smart Grid: a

Review. Smart Grid Conf., 2013, p. 96–101. doi:10.1109/SGC.2013.6733807.

[63] McDaniel P, McLaughlin S. Security and privacy challenges in the smart grid. IEEE

Secur Priv 2009;7:75–7. doi:10.1109/MSP.2009.76.

[64] Dawoud B, Amer E, Gross D. Experimental investigation of an adsorptive thermal

energy storage. Int J Energy Res 2007;31:135–47. doi:10.1002/er.

[65] Parikh PP, Kanabar MG, Sidhu TS. Opportunities and challenges of wireless

communication technologies for smart grid applications. IEEE PES Gen Meet 2010:1–

7. doi:10.1109/PES.2010.5589988.

[66] Fan Z, Kulkarni P, Gormus S, Efthymiou C, Kalogridis G, Sooriyabandara M, et al.

Smart grid communications: Overview of research challenges, solutions, and

standardization activities. IEEE Commun Surv Tutorials 2013;15:21–38.

doi:10.1109/SURV.2011.122211.00021.

[67] Souryal MR, Golmie N. Analysis of advanced metering over a wide area cellular

network. 2011 IEEE Int Conf Smart Grid Commun SmartGridComm 2011 2011:102–

7. doi:10.1109/SmartGridComm.2011.6102299.

Page | 89

[68] Iyer G, Agrawal P, Monnerie E, Cardozo RS. Performance analysis of wireless mesh

routing protocols for smart utility networks. 2011 IEEE Int Conf Smart Grid Commun

SmartGridComm 2011 2011:114–9. doi:10.1109/SmartGridComm.2011.6102301.

[69] Kuzlu M, Pipattanasompom M, Rahman S. A comprehensive review of smart grid

related standards and protocols. ICSG 2017 - 5th Int Istanbul Smart Grids Cities Congr

Fair 2017:12–6. doi:10.1109/SGCF.2017.7947600.

[70] MacKiewicz RE. Overview of IEC 61850 and benefits. 2006 IEEE PES Power Syst

Conf Expo PSCE 2006 - Proc 2006;57:623–30. doi:10.1109/PSCE.2006.296392.

[71] Society IPE. IEEE Standard for SCADA and Automation Systems. vol. 2008. 2008.

[72] Basso TS, DeBlasio R. IEEE 1547 series of standards: Interconnection issues. IEEE

Trans Power Electron 2004;19:1159–62. doi:10.1109/TPEL.2004.834000.

[73] (UCAIug) A-STF. Security Profile for Advanced Metering Infrastructure. 2012.

[74] Cleveland F. IEC TC57 security standards for the power system’s information

infrastructure - Beyond simple encryption. Proc IEEE Power Eng Soc Transm Distrib

Conf 2006:1079–87. doi:10.1109/TDC.2006.1668652.

[75] Vuppala SK, Member S, Padmanabh K, Bose SK. Incorporating Fairness within

Demand Response Programs in Smart Grid. Innov. Smart Grid Technol., 2011, p. 1–9.

[76] Kefayati M, Baldick R. Energy Delivery Transaction Pricing for Flexible Electrical

Loads. IEEE Int. Conf. Smart Grid Commun., 2011, p. 363–8.

[77] Baharlouei Z, Narimani H, Mohsenian-rad H. Tackling Co-existence and Fairness

Challenges in Autonomous Demand Side Management. Glob. Commun. Conf., 2012,

p. 3159–64.

[78] Shinwari M, Youssef A, Hamouda W. A Water-Filling Based Scheduling Algorithm

for the Smart Grid. IEEE Trans Smart Grid 2012;3:710–9.

[79] Baharlouei Z, Hashemi M, Narimani H, Mohsenian-rad H. Achieving Optimality and

Fairness in Autonomous Demand Response : Benchmarks and Billing Mechanisms.

IEEE Trans Smart Grid 2013;4:968–75.

Page | 90

[80] Güngör VC, Sahin D, Kocak T, Ergüt S, Buccella C, Cecati C, et al. Smart grid

technologies: Communication technologies and standards. IEEE Trans Ind Informatics

2011;7:529–39. doi:10.1109/TII.2011.2166794.

[81] Hossain E, Han Z, Poor HV. Smart grid communications and networking. Cambridge

University Press; 2012.

[82] Office of Electricity | Department of Energy 2018. https://www.energy.gov/oe/office-

electricity (accessed August 28, 2018).

[83] Yan Y, Qian Y, Sharif H, Tipper D. A Survey on Smart Grid Communication

Infrastructures : Motivations , Requirements and Challenges A Survey on Smart Grid

Communication Infrastructures : Motivations , Requirements and Challenges. IEEE

Commun Surv Tutorials 2013;15:5–20. doi:10.1109/SURV.2012.021312.00034.

[84] Erol-Kantarci M, Mouftah HT. Wireless multimedia sensor and actor networks for the

next generation power grid. Ad Hoc Networks 2011;9:542–51.

doi:10.1016/j.adhoc.2010.08.005.

[85] Jayadev V, Swarup KS. Optimization of Microgrid with Demand Side Management

using Genetic Algorithm Indian Institute of Technology Madras , India. IET Conf.

Power Unity a Whole Syst. Approach, 2013, p. 1–12.

[86] Bharathi C, Rekha D, Vijayakumar V. Genetic Algorithm Based Demand Side

Management for Smart Grid. Wirel Pers Commun 2017;93:481–502.

doi:10.1007/s11277-017-3959-z.

[87] Connor J, Seyedmahmoudian M, Horan B. Using particle swarm optimization for PID

optimization for altitude control on a quadrotor. Australas. Univ. Power Eng. Conf.,

2018, p. 1–6. doi:10.1109/AUPEC.2017.8282423.

[88] Logenthiran T, Srinivasan D, Phyu E. Particle swarm optimization for demand side

management in smart grid. IEEE Innov. Smart Grid Technol., IEEE; 2015, p. 1–6.

doi:10.1109/ISGT-Asia.2015.7386973.

[89] Nayak SK, Sahoo NC, Panda G. Demand Side Management of Residential Loads in a

Smart Grid using 2D Particle Swarm Optimization Technique. IEEE Power, Commun.

Inf. Technol. Conf., 2015, p. 201–6.

Page | 91

[90] Bozorg-Haddad O, Solgi M, Loáiciga HA. Meta-Heuristic and Evolutionary

Algorithms for Engineering Optimization. John Wiley & Sons, Inc.; 2017.

doi:10.1002/9781119387053.

[91] Man KF, Tang KS, Kwong S. Genetic algorithms: Concepts and applications. IEEE

Trans Ind Electron 1996;43:519–34. doi:10.1109/41.538609.

[92] Back T. Selective pressure in evolutionary algorithms: a characterization of selection

mechanisms 2002:57–62. doi:10.1109/icec.1994.350042.

[93] Appl J, Valenzuela F. Scheduling electric power generators using particle swarm

optimization combined with the lagrangian relaxation method. Int J Appl Math

Comput Sci 2004;14:411–21.

[94] Chuanwen J, Bompard E. A self-adaptive chaotic particle swarm algorithm for short

term hydroelectric system scheduling in deregulated environment. Energy Convers

Manag 2005;46:2689–96. doi:10.1016/j.enconman.2005.01.002.

[95] Gaur S, Ch S, Graillot D, Chahar BR, Kumar DN. Application of Artificial Neural

Networks and Particle Swarm Optimization for the Management of Groundwater

Resources. Water Resour Manag 2013;27:927–41. doi:10.1007/s11269-012-0226-7.

[96] Saadatpour M, Afshar A. Multi Objective Simulation-Optimization Approach in

Pollution Spill Response Management Model in Reservoirs. Water Resour Manag

2013;27:1851–65. doi:10.1007/s11269-012-0230-y.

[97] Afshar A, Massoumi F, Afshar A, Mariño MA. State of the Art Review of Ant Colony

Optimization Applications in Water Resource Management. Water Resour Manag

2015;29:3891–904. doi:10.1007/s11269-015-1016-9.

[98] Christodoulou S. Scheduling Resource-Constrained Projects with Ant Colony

Optimization Artificial Agents. J Comput Civ Eng 2009;24:45–55.

doi:10.1061/(asce)0887-3801(2010)24:1(45).

[99] Merkle D, Middendorf M, Schmeck H. Ant colony optimization for resource-

constrained project scheduling. IEEE Trans Evol Comput 2002;6:333–46.

doi:10.1109/TEVC.2002.802450.

Page | 92

[100] Dorigo M, Maniezzo V, Colorni A. Ant system: Optimization by a colony of

cooperating agents. IEEE Trans Syst Man, Cybern Part B Cybern 1996;26:29–41.

doi:10.1109/3477.484436.

[101] Vanderbei RJ. Linear Programming: Foundations and Extensions. Springer; 1997.

[102] Karlin S. Matrix Games, Programming, and Mathematical Economics: Mathematical

Methods and Theory in Games, Programming, and Economics. Addison-Wesley

Publishing Company, Elsevier Inc.; 1959. doi:10.1016/B978-1-4831-9795-1.50009-7.

[103] Bitran GR, Novaes AG. Linear Programming with a Fractional Objective Function.

Oper Res 1973;21:22–9. doi:10.2307/169085.

[104] Hillier FS, Lieberman GJ. Introduction to operations research. McGraw-Hill Higher

Education; 2010.

[105] Gill PE, Murray W, Saunders MA, Wright MH. Constrained nonlinear programming.

Handbooks Oper. Res. Manag. Sci., North-Holland: Elsevier Science Publishers B.V.;

1989, p. 171–210. doi:10.1016/s0927-0507(89)01004-2.

[106] Zangwill WI. Nonlinear programming : a unified approach. Englewood Cliffs, N.J. :

Prentice-Hall; 1969.

[107] Vanderbei RJ, Shanno DF. An interior-point method for nonconvex nonlinear

programming. Comput Optim Appl 1999;13:231–52.

[108] Hadley GF. Nonlinear and Dynamic Programming. Econometrica 1966;34:248–9.

[109] Richard E. Bellman and Stuart E. Dreyfus. Applied dynamic programming. Princeton

university press; 2015.

[110] Aghajani GR, Shayanfar HA, Shayeghi H. Demand side management in a smart

micro-grid in the presence of renewable generation and demand response. Energy

2017;126:622–37. doi:10.1016/j.energy.2017.03.051.

[111] Mohsenian-Rad AH, Wong VWS, Jatskevich J, Schober R. Optimal and autonomous

incentive-based energy consumption scheduling algorithm for smart grid. Innov. Smart

Grid Technol. Conf., 2010, p. 1–6. doi:10.1109/ISGT.2010.5434752.

Page | 93

[112] Wang K, Li H, Maharjan S, Zhang Y, Guo S. Green Energy Scheduling for Demand

Side Management in the Smart Grid. IEEE Trans Green Commun Netw 2018;2:596–

611. doi:10.1109/TGCN.2018.2797533.

[113] Samadi P, Mohsenian-Rad H, Schober R, Wong VWS. Advanced Demand Side

Management for the Future Smart Grid Using Mechanism Design. IEEE Trans Smart

Grid 2012;3:1170–80. doi:10.1109/TSG.2012.2203341.

[114] Hyoungchel Bae, Jongha Yoon, Yunseong Lee, Juho Lee, Taejin Kim, Jeongseok Yu,

et al. User-friendly demand side management for smart grid networks. Int. Conf. Inf.

Netw., 2014, p. 481–5. doi:10.1109/ICOIN.2014.6799728.

[115] Mohsenian-Rad AH, Wong VWS, Jatskevich J, Schober R, Leon-Garcia A.

Autonomous demand-side management based on game-theoretic energy consumption

scheduling for the future smart grid. IEEE Trans Smart Grid 2010;1:320–31.

doi:10.1109/TSG.2010.2089069.

[116] Wang J, Sun Z, Zhou Y, Dai J. Optimal dispatching model of smart home energy

management system. IEEE Innov. Smart Grid Technol., 2012, p. 1–5.

doi:10.1109/ISGT-Asia.2012.6303266.

[117] Liu B, Kang J, Jiang N, Jing Y. Cost Control of the Transmission Congestion

Management in Electricity Systems Based on Ant Colony Algorithm. Energy Power

Eng 2011;03:17–23. doi:10.4236/epe.2011.31003.

[118] Mahmood A, Ullah MN, Razzaq S, Basit A, Mustafa U, Naeem M, et al. A new

scheme for demand side management in future smart grid networks. Procedia Comput

Sci 2014;32:477–84. doi:10.1016/j.procs.2014.05.450.

[119] Nguyen HK, Song J Bin, Han Z. Distributed Demand Side Management with Energy

Storage in Smart Grid. IEEE Trans Parallel Distrib Syst 2015;26:3346–57.

doi:10.1109/TPDS.2014.2372781.

[120] Kennedy J, Eberhart RC. A discrete binary version of the particle swarm algorithm.

IEEE Int. Conf. Comput. Cybern. Simul., 1997, p. 4104–8.

[121] Pedrasa MAA, Member S, Spooner TD, Macgill IF. Scheduling of Demand Side

Resources Using Binary Particle Swarm Optimization. IEEE Trans Power Syst

2009;24:1173–81.

Page | 94

[122] Mahmood D, Javaid N, Alrajeh N, Khan ZA, Qasim U, Ahmed I, et al. Realistic

Scheduling Mechanism for smart homes. Energies 2016;9:1–28.

doi:10.3390/en9030202.

[123] Zhou Y, Chen Y, Xu G, Zhang Q, Krundel L. Home Energy Management with PSO in

Smart Grid. IEEE 23rd Int. Symp. Ind. Electron., 2014, p. 1666–70.

[124] Zhou Y, Xu G. Demand side energy management with PSO and regulated electric

vehicles behaviours. IEEE PES Asia-Pacific Power Energy Eng. Conf., 2014, p. 1–6.

doi:10.1109/APPEEC.2014.7066103.

[125] Dethlefs T, Preisler T, Renz W. Ant-Colony based Self-Optimization for Demand-

Side-Management. Smarter Eur., 2015, p. 1–8.

[126] Rahim S, Iqbal Z, Shaheen N, Khan ZA, Qasim U, Khan SA, et al. Ant Colony

Optimization Based Energy Management Controller for Smart Grid. IEEE 30th Int.

Conf. Adv. Inf. Netw. Appl., 2016, p. 1154–9. doi:10.1109/AINA.2016.163.

[127] Hazra J, Das K, Seetharam DP. Smart grid congestion management through demand

response. IEEE Third Int. Conf. Smart Grid Commun., 2012, p. 109–14.

doi:10.1109/SmartGridComm.2012.6485968.

[128] Okonta CI, Kemp AH, Edopkia RO, Monyei GC, Okelue ED. A heuristic based ant

colony optimization algorithm for energy efficient smart homes. 5th Int. Conf. Exhib.

Clean Energy, 2016, p. 1–12.

[129] Kapade N, Deshmukh S. Comparative study of game theoretic approaches in

distributed system. World Congr. Comput. Inf. Technol., 2013, p. 1–4.

doi:10.1109/WCCIT.2013.6618672.

[130] Deng R, Member S, Yang Z, Chen J, Member S. Residential Energy Consumption

Scheduling : A Coupled-Constraint Game Approach. IEEE Trans Smart Grid

2014;5:1340–50. doi:10.1109/TSG.2013.2287494.

[131] Mohsenian-Rad A-H, Wong VWS, Jatskevich J, Schober R, Leon-Garcia A.

Autonomous Demand Side Management Based on Game-Theoretic Energy

Consumption Scheduling for the Future Smart Grid. IEEE Trans Smart Grid

2010;1:320–31. doi:10.1109/TSG.2010.2089069.

Page | 95

[132] Kurucz CN, Brandt D, Sim S. Linear Programming Model for Reducing System Peak

Through Customer Load Control Programs. IEEE Trans Power Syst 1996;11:1817–24.

doi:10.1109/59.544648.

[133] Zhu Z, Tang J, Lambotharan S, Chin WH, Fan Z. An Integer Linear Programming

Based Optimization for Home Demand-side Management in Smart Grid. IEEE PES

Innov. Smart Grid Technol., 2012, p. 1–5. doi:10.1109/ISGT.2012.6175785.

[134] Martins AG, Coelho D, Antunes CH, Clímaco J. A Multiple Objective Linear

Programming Approach to Power Generation Planning with Demand-Side

Management (DSM). Int Trans Oper Res 1996;3:305–17. doi:10.1111/j.1475-

3995.1996.tb00055.x.

[135] Kontorinaki M, Spiliopoulou A, Papamichail I, Papageorgiou M, Tyrinopoulos Y,

Chrysoulakis J. Overview of nonlinear programming methods suitable for calibration

of traffic flow models. Oper Res 2015;15:327–36. doi:10.1007/s12351-014-0146-9.

[136] Shaaban MF, Osman AH, Hassan MS. Day-ahead Optimal Scheduling for Demand

Side Management in Smart Grids. Eur. Model. Symp., 2016, p. 124–9.

doi:10.1109/EMS.2016.30.

[137] Helal SA, Najee RJ, Hanna MO, Shaaban MF, Osman AH, Hassan MS. An energy

management system for hybrid microgrids in remote communities. IEEE 30th Can

Conf Electr Comput Eng 2017:1–4. doi:10.1109/CCECE.2017.7946775.

[138] Chu WC, Chen BK, Fu CK. Dispatch of direct load control using dynamic

programming. IEEE Trans Power Syst 1993;8:1525–30. doi:10.1109/59.260955.

[139] Hsu Y, Member S. Dispatch of Direct Load Control Using Dynamic Programming.

IEEE Trans Power Syst 1991;6:1056–61. doi:10.1109/59.119246.

[140] Reka SS, Ramesh V. Demand side management scheme in smart grid with cloud

computing approach using stochastic dynamic programming. Perspect Sci 2016;8:169–

71. doi:10.1016/j.pisc.2016.04.024.

[141] Pina A, Silva C, Ferrão P. The impact of demand side management strategies in the

penetration of renewable electricity. Energy 2012;41:128–37.

doi:10.1016/j.energy.2011.06.013.

Page | 96

[142] Gabash A, Li P. Flexible optimal operation of battery storage systems for energy

supply networks. IEEE Trans Power Syst 2013;28:2788–97.

doi:10.1109/TPWRS.2012.2230277.

[143] Kanchev H, Lu D, Colas F, Lazarov V, Francois B, Member S. Energy Management

and Operational Planning of a Microgrid With a PV-Based Active Generator for Smart

Grid Applications. IEEE Trans Ind Electron 2011;58:4583–92.

doi:10.1109/TIE.2011.2119451.

[144] Quiggin D, Cornell S, Tierney M, Buswell R. A simulation and optimisation study:

Towards a decentralised microgrid, using real world fluctuation data. Energy

2012;41:549–59. doi:10.1016/j.energy.2012.02.007.

[145] Dietrich K, Latorre JM, Olmos L, Ramos A. Demand response in an isolated system

with high wind integration. IEEE Trans Power Syst 2012;27:20–9.

doi:10.1109/TPWRS.2011.2159252.

[146] Çiçek N, Deliç H. Demand Response Management for Smart Grids With Wind Power.

IEEE Trans Sustain Energy 2015;6:625–34. doi:10.1109/TSTE.2015.2403134.

[147] Amrollahi MH, Bathaee SMT. Techno-economic optimization of hybrid

photovoltaic/wind generation together with energy storage system in a stand-alone

micro-grid subjected to demand response. Appl Energy 2017;202:66–77.

doi:10.1016/j.apenergy.2017.05.116.

[148] Wang X, Palazoglu A, El-Farra NH. Operational optimization and demand response of

hybrid renewable energy systems. Appl Energy 2015;143:324–35.

doi:10.1016/j.apenergy.2015.01.004.

[149] Behboodi S, Chassin DP, Djilali N, Crawford C. Interconnection-wide hour-ahead

scheduling in the presence of intermittent renewables and demand response: A surplus

maximizing approach. Appl Energy 2017;189:336–51.

doi:10.1016/j.apenergy.2016.12.052.

[150] Wu Z, Tazvinga H, Xia X. Demand side management of photovoltaic-battery hybrid

system. Appl Energy 2015;148:294–304. doi:10.1016/j.apenergy.2015.03.109.

[151] Blake ST, Sullivan DTJO. Optimization of distributed energy resources in an industrial

microgrid. Procedia CIRP 2018;67:104–9. doi:10.1016/j.procir.2017.12.184.

Page | 97

[152] Palma-Behnke R, Benavides C, Aranda E, Llanos J, Sáez D. Energy Management

System for a Renewable based Microgrid with a Demand Side Management

Mechanism. IEEE Symp. Comput. Intell. Appl. Smart Grid, 2011, p. 1–8.

doi:10.1109/CIASG.2011.5953338.

[153] Fan W, Liu N, Zhang J. An Event-Triggered Online Energy Management Algorithm of

Smart Home: Lyapunov Optimization Approach. Energies 2016;9:381.

doi:10.3390/en9050381.

[154] Yang P, Chavali P, Nehorai A. Parallel autonomous optimization of demand response

with renewable distributed generators. 2012 IEEE 3rd Int Conf Smart Grid Commun

SmartGridComm 2012 2012:55–60. doi:10.1109/SmartGridComm.2012.6485959.

[155] Shen J, Jiang C, Liu Y, Qian J. A Microgrid Energy Management System with

Demand Response for Providing Grid Peak Shaving. Electr Power Components Syst

2016;44:843–52. doi:10.1080/15325008.2016.1138344.

[156] Solanki B V., Raghurajan A, Bhattacharya K, Canizares CA. Including Smart Loads

for Optimal Demand Response in Integrated Energy Management Systems for Isolated

Microgrids. IEEE Trans Smart Grid 2017;8:1739–48. doi:10.1109/TSG.2015.2506152.

[157] Nikolas Philippou, Venizelou V, Makrides G, Hadjipanayi M, Efthymiou V,

Georghiou GE. PV integration and price-based Demand Side Management: Optimum

time-of-use tariffs. Proc 33th EUPVSEC 2016.

[158] Erol-Kantarci M, Mouftah HT. Using wireless sensor networks for energy-aware

homes in smart grids. IEEE Symp. Comput. Commun., 2010, p. 456–8.

doi:10.1109/ISCC.2010.5546765.

[159] Gudi N, Wang L, Devabhaktuni V. A demand side management based simulation platform

incorporating heuristic optimization for management of household appliances. Int J Electr Power

Energy Syst 2012;43:185–93. doi:10.1016/j.ijepes.2012.05.023.

[160] Hsu Y, Member S. DISPATCH OF DIRECT LOAD CONTROL USING DYNAMIC

PROGRAMMING. Control 1991;6:1056–61.

[161] Anvari-Moghaddam A, Monsef H, Rahimi-Kian A. Optimal smart home energy

management considering energy saving and a comfortable lifestyle. IEEE Trans Smart

Grid 2015;6:324–32. doi:10.1109/TSG.2014.2349352.

Page | 98

[162] Muralitharan K, Sakthivel R, Shi Y. Neurocomputing Multiobjective optimization

technique for demand side management with load balancing approach in smart grid.

Neurocomputing 2016;177:110–9. doi:10.1016/j.neucom.2015.11.015.

[163] Shakouri G. H, Kazemi A. Multi-objective cost-load optimization for demand side

management of a residential area in smart grids. Sustain Cities Soc 2017;32:171–80.

doi:10.1016/j.scs.2017.03.018.

[164] Jalali MM, Kazemi A. Demand side management in a smart grid with multiple

electricity suppliers. Energy 2015;81:766–76. doi:10.1016/j.energy.2015.01.027.

[165] Manzoor A, Javaid N, Ullah I, Abdul W, Almogren A, Alamri A. An intelligent hybrid

heuristic scheme for smart metering based demand side management in smart homes.

Energies 2017;10:1–28. doi:10.3390/en10091258.

[166] del Valle Y, Venayagamoorthy GK, Mohagheghi S, Hernandez JC, Harley RG.

Particle swarm optimization: Basic concepts, variants and applications in power

systems. IEEE Trans Evol Comput 2008;12:171–95. doi:10.1109/TEVC.2007.896686.

[167] Halder PK, Paul N, Beg MRA. Prospect of Pongamia pinnata (Karanja) in Bangladesh:

A Sustainable Source of Liquid Fuel. J Renew Energy 2014:1–12.

[168] Tazi K, Abbou FM, Chaka AB, Abdi F. Modeling and simulation of a residential

microgrid supplied with PV/batteries in connected/disconnected modes — Case of

Morocco. J Renew Sustain Energy 2017;9:1–15. doi:10.1063/1.4979355.

[169] Zunnurain I, Maruf M, Rahman M, Shafiullah G. Implementation of Advanced

Demand Side Management for Microgrid Incorporating Demand Response and Home

Energy Management System. Infrastructures 2018;3:50.

doi:10.3390/infrastructures3040050.

[170] Halder P, Azad K, Shah S, Sarker E. Prospects and technological advancement of

cellulosic bioethanol ecofuel production. In: Azad K, editor. Adv. Eco-Fuels a Sustain.

Environ., 2019, p. 211–36.

[171] Osório GJ, Shafie-khah M, Lotfi M, Ferreira-Silva BJM, Catalão JPS. Demand-Side

Management of Smart Distribution Grids Incorporating Renewable Energy Sources.

Energies 2019;12:143. doi:10.3390/en12010143.

Page | 99

[172] Nazir R, Laksono HD, Waldi EP, Ekaputra E, Coveria P. Renewable energy sources

optimization: A micro-grid model design. Energy Procedia 2014;52:316–27.

doi:10.1016/j.egypro.2014.07.083.

[173] Longe OM, Ouahada K, Rimer S, Harutyunyan AN, Ferreira HC. Distributed demand

side management with battery storage for smart home energy scheduling. Sustain

2017;9. doi:10.3390/su9010120.

[174] Olivares DE, Mehrizi-Sani A, Etemadi AH, Cañizares CA, Iravani R, Kazerani M, et

al. Trends in microgrid control. IEEE Trans Smart Grid 2014;5:1905–19.

doi:10.1109/TSG.2013.2295514.

[175] Behrangrad M. A review of demand side management business models in the electricity

market. Renew Sustain Energy Rev 2015;47:270–83. doi:10.1016/j.rser.2015.03.033.

[176] Siano P. Demand response and smart grids - A survey. Renew Sustain Energy Rev

2014;30:461–78. doi:10.1016/j.rser.2013.10.022.

[177] Ferruzzi G, Graditi G, Rossi F, Russo A. Optimal Operation of a Residential

Microgrid: The Role of Demand Side Management. Intell Ind Syst 2015;1:61–82.

doi:10.1007/s40903-015-0012-y.

[178] Tascikaraoglu A, Boynuegri AR, Uzunoglu M. A demand side management strategy

based on forecasting of residential renewable sources : A smart home system in

Turkey. Energy Build 2014;80:309–20. doi:10.1016/j.enbuild.2014.05.042.

[179] Gottwalt S, Ketter W, Block C, Collins J, Weinhardt C. Demand side management-A

simulation of household behavior under variable prices. Energy Policy 2011;39:8163–

74. doi:10.1016/j.enpol.2011.10.016.

[180] Ma K, Yao T, Yang J, Guan X. Electrical Power and Energy Systems Residential

power scheduling for demand response in smart grid. Int J Electr Power Energy Syst

2016;78:320–5. doi:10.1016/j.ijepes.2015.11.099.

[181] Özkan HA. Appliance based control for Home Power Management Systems. Energy

2016;114:693–707. doi:10.1016/j.energy.2016.08.016.

[182] Ogunjuyigbe ASO, Ayodele TR, Akinola OA. User satisfaction-induced demand side

load management in residential buildings with user budget constraint. Appl Energy

2017;187:352–66. doi:10.1016/j.apenergy.2016.11.071.

Page | 100

[183] Steen D, Tuan LA, Carlson O. Effects of Network Tariffs on Residential Distribution

Systems and Price-Responsive Customers Under Hourly Electricity Pricing. IEEE

Trans Smart Grid 2016;7:617–26. doi:10.1109/TSG.2015.2464789.

[184] Missaoui R, Joumaa H, Ploix S, Bacha S. Managing energy Smart Homes according to

energy prices: Analysis of a Building Energy Management System. Energy Build

2014;71:155–67. doi:10.1016/j.enbuild.2013.12.018.

[185] Lu Y, Wang S, Sun Y, Yan C. Optimal scheduling of buildings with energy generation

and thermal energy storage under dynamic electricity pricing using mixed-integer

nonlinear programming. Appl Energy 2015;147:49–58.

doi:10.1016/j.apenergy.2015.02.060.

[186] Sarker E, Mekhilef S, Stojcevski A. Optimal scheduling of appliances in smart grid

environment using BPSO algorithm. 11th Int. Conf. Appl. Energy, 2019, p. 1–4.

[187] Shen J, Jiang C, Liu Y, Qian J. A Microgrid Energy Management System with

Demand Response for Providing Grid Peak Shaving. Electr Power Components Syst

2016;44:843–52. doi:10.1080/15325008.2016.1138344.

[188] Kumar Nunna HSVS, Doolla S. Energy management in microgrids using demand

response and distributed storage - A multiagent approach. IEEE Trans Power Deliv

2013;28:939–47. doi:10.1109/TPWRD.2013.2239665.

[189] Gonzalez de Durana J, Barambones O. Technology-free microgrid modeling with

application to demand side management. Appl Energy 2018;219:165–78.

doi:10.1016/j.apenergy.2018.03.024.

[190] Akram U, Khalid M, Shafiq S. Optimal Sizing of a Wind/Solar/Battery Hybrid Grid-

connected Microgrid Optimal sizing of a wind/solar/battery hybrid grid-connected

microgrid system. IET Renew Power Gener Res 2018;12:72–80. doi:10.1049/iet-

rpg.2017.0010.

[191] Roy S. Market constrained optimal planning for wind energy conversion systems over

multiple installation sites. IEEE Trans Energy Convers 2002;17:124–9.

doi:10.1109/60.986449.

Page | 101

Appendix

Table A 1 Average hourly load consumption in weekday

Time (h) Load (kWh)

Household1 Household 2 Household 3 Household 4

0:00 0.56 0.31 0.12 0.11

1:00 0.64 0.22 0.09 0.11

2:00 0.11 0.09 0.1 0.17

3:00 0.09 0.09 0.1 0.21

4:00 0.12 0.1 0.5 0.29

5:00 0.09 0.16 0.1 0.35

6:00 0.11 0.14 0.1 0.49

7:00 0.25 0.09 0.86 0.59

8:00 0.46 0.08 0.09 0.66

9:00 0.12 0.86 0.16 0.71

10:00 0.15 1.03 0.19 0.58

11:00 0.48 0.83 1.36 0.68

12:00 0.71 0.91 0.79 0.51

13:00 1.01 0.65 1.64 0.95

14:00 0.68 0.87 1.08 0.81

15:00 2.18 0.93 0.09 1.48

16:00 1.08 1.57 0.61 0.73

17:00 1.13 1.06 1.44 1.05

18:00 1.06 1.09 1.24 0.88

19:00 0.98 0.91 1.11 0.55

20:00 1.09 0.79 0.84 0.63

21:00 0.72 0.81 0.48 0.56

22:00 0.61 0.16 0.14 0.51

23:00 0.53 0.13 0.13 0.51

Page | 102

Table A 2 Average hourly load consumption in weekend

Time (h) Load (kWh)

Household1 Household 2 Household 3 Household 4

0:00 0.11 0.16 0.12 0.12

1:00 0.12 0.16 0.1 0.11

2:00 0.56 0.13 0.1 0.13

3:00 0.11 0.16 0.08 0.11

4:00 0.12 0.16 0.08 0.11

5:00 0.58 0.16 0.09 0.16

6:00 0.11 0.75 0.1 0.1

7:00 0.1 0.14 0.54 0.1

8:00 0.35 0.14 0.27 1.06

9:00 0.96 1.01 1.51 1.46

10:00 1.1 0.21 2.48 1.83

11:00 0.29 0.11 1.09 0.74

12:00 0.11 0.58 1.77 1.7

13:00 0.61 0.15 0.14 1.31

14:00 0.65 0.87 0.91 2.13

15:00 2.46 1.74 0.58 0.51

16:00 1.38 2.54 0.11 0.29

17:00 1.22 2.89 0.61 0.79

18:00 2.79 0.85 0.83 1.03

19:00 0.29 0.16 1.6 0.98

20:00 0.82 0.65 0.81 0.11

21:00 0.66 0.16 0.78 0.65

22:00 0.28 0.32 0.13 0.13

23:00 0.61 0.45 0.19 0.13

Documents

Modeling and optimization of Demand side Management in ......v List of Publications 1. Eity Sarker, Pobitra Halder, Mehdi Seyedmahmoudian, Elmira Jamei, Ben Horan, Saad Mekhilef, Alex