Transportation Research Part D 62 (2018) 386–417

Contents lists available at ScienceDirect

Transportation Research Part D

journal homepage: www.elsevier.com/locate/trd

A review of EVs charging: From the perspective of energy

T
optimization, optimization approaches, and charging techniques
⁎
Muhammad Amjada, , Ayaz Ahmadb, Mubashir Husain Rehmanic, Tariq Umerd
a
School of Computer Science and Electronic Engineering, University of Essex, UK
b
Department of Electrical Engineering, COMSATS Institute of Information Technology, Wah Cantt., Pakistan
c
Waterford Institute of Technology, Ireland
d
Department of Computer Science, COMSATS Institute of Information Technology, Wah Cantt., Pakistan

A R T IC LE I N F O ABS TRA CT

Keywords: Introduction of electric vehicles (EVs) or plug-in electric vehicles (PEVs) in the road transpor-
EV charging tation can significantly reduce the carbon emission. Hence, the demand of EVs is likely to in-
Battery life crease in the near future. Large penetration of EVs will also ultimately result into high loads on
Optimization methods the existing power grids. The controlled charging of EVs can have a significant impact on the
Optimization objectives
power grid load, voltage, frequency, and power losses. In this paper, we have provided a com-
Centralized EVs charging
Distributed EVs charging
prehensive review of various energy optimization approaches used for EVs charging. Energy
optimization approaches used for EVs not only enhance the battery life but also contribute in
regulating the voltage and frequency. During EVs charging, various objective functions such as
supporting the renewable energy sources, minimization of the peak load, energy cost, and
maximization of the aggregator profit have also been studied from optimization perspectives. The
controlled and an optimized EVs charging enhances the performance of EVs batteries and con-
serves the energy in the system by minimizing the load and power losses. The different EVs
charging approaches such as centralized and distributed suited for different objective functions
have also been studied and compared with respect to various optimization approaches.

1. Introduction

1.1. Motivation and background

Electric Vehicles (EVs) or plug-in hybrid electric vehicles (PHEV) have gained much attention in today’s modern world due to
their economic and environmental gains. It is estimated that major portion of oil is consumed by the transporting machinery. This
huge utilization of oil results into drastic effects such as climate change, environmental degradation, and economic ups and downs.
An EV can be defined as any light weight vehicle which can operate with chargeable battery of 4 kW h. EVs have following benefits as
compared to other internal combustion engine vehicles (ICEV) (García-Villalobos et al., 2014).

• EVs minimize the CO emission in environment. This benefit of EVs improves the environment and has a positive effect on climate.
2

• Less CO emission results in less pollution in our cities. Healthier atmosphere of cities improves the overall health and standard of
2
citizens.

⁎
Corresponding author.
E-mail addresses: [email protected] (M. Amjad), [email protected] (A. Ahmad), [email protected] (M.H. Rehmani),
[email protected] (T. Umer).

https://doi.org/10.1016/j.trd.2018.03.006

Available online 29 March 2018

1361-9209/ © 2018 Elsevier Ltd. All rights reserved.
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

• Compared to ICEV, EVs are cost-effective solution for transportation. Transportation efficiency can increase up to 83 times with
the inclusion of EVs on the roads.
• EVs can be charged from various locations. They can be charged even from the home using home plug.
• EVs charging also favors the use of renewable energy sources (Seddig et al., 2017). Therefore, electric grids efficiency and output
can be enhanced with the integration of EVs into power structure.

Less fuel consumption cuts the carbon emission which comes out to be environmental friendly. Inclusion of EVs enhances energy
efficiency, minimizes the greenhouse gases and foreign oil importation, but it has also introduced a burden on power grid. However,
this burden on the existing power sources can be minimized with the controlled and coordinated charging of EVs. This can be
achieved by employing the efficient charging infrastructure. The charging of EVs can also be regulated with the help of different
optimization approaches. The optimization approaches are also helpful in achieving the various objectives during the EVs charging
methods. These objectives can be the minimization in power losses, provision of ancillary services, integration of renewable energy
sources, and regulating the voltages.
Therefore, EVs charging is studied from various perspectives for minimizing the load on smart grids. It is expected that EVs will
grow in a larger number in next decade. Uncontrolled or un-coordinated charging of EVs will not only degrade the power grid
capability but will also affect the distribution facility of the overall power structure. If the electrification of transportation is not
properly optimized, this will ultimately lead to the collapse of existing power structure (Clement et al., 2009a; Amjad et al., in press;
Umer et al., 2016b).
For optimal charging of EVs, a well-coordinated and controlled infrastructure is needed. EVs charging is now well supported with
the help of smart grid. Amplification of peak demand and voltage fluctuations, demand the smart charging approaches to control the
electrification of EVs. It is estimated by various studies, that only 10% integration of EVs into the roads can cause a serious blow to
the existing power structure. Originally power grids are designed to support the commercial and house hold utilities. Interest in smart
grids for smart charging of EVs is replacing the existing grid. Smart charging is dealing with challenges like stability, infrastructure
modifications, and load management. Following are the characteristics of smart grid (Habib et al., 2015; Fang et al., 2012; Mouftah
et al., 2017; Rehan et al., 2017).

• Enhancement of power quality.

• Significant improvement in electric power generation and distribution system.
• Introduction of robustness in the system.
• Efficient handling of EVs and their charging.
• Maximizing the grid security.
• Supportive role for integration of renewable energy sources.
For charging EVs, information regarding its charging needs is collected. This information is harnessed for optimal charging of EVs.
Power from electric grids are transferred and then stored in EVs during charging state, and energy is usually consumed during
discharging. EVs can enhance the load on distribution network and could also contribute as power generator to distribution network.
The charging capability of EVs can be studied from various perspectives. Usually, the following factors are involved during the
charging of EVs.

• Charging duration: How much time and at what time EVs fleet can be charged. The amount of energy consumption by EVs
determine the charging duration. EVs owners point of view also does matter in deciding the charging duration. Some EVs have to
cover a large distance, some have to be used as a power source at peak load, and others have their local generation capability. All
these factors are taken into account for optimal decision of charging duration.
• EVs mobility or parking pattern:Mobility models affect the charging of EVs. How much to charged and when to charge can be
decided if the mobility model and parking pattern is used as the deciding factor for charging of EVs.
• Charging Approach: EVs can be charged using different approaches. In most cases, centralized and distributed approaches are
used.
• Unidirectional or bidirectional connectivity: Some EVs which act as power source and have local generation capability,
contribute to electric grid and follows the bidirectional connectivity to charging stations. EVs which only get charged from
charging locations used the unidirectional connectivity.
• Type and number of EVs charging at a given location:EVs charging have a significant impact on local power grids. The
charging of EVs fleet during a peak load of distribution network can lead to the degradation of voltage. Hence maximum charging
is done during peak time to avoid any bottleneck in distribution network and to enhance the reliability of electric grid.
• EVs batteries charging and discharging states: State of charge (SOC) of batteries is estimated for scheduling the charging of
EVs. EVs batteries should be in such state, that the information regarding their current level of charging and remaining power can
be communicated to the central charging entity.
• Current and voltage requirements:During peak hours, the voltage and current are more susceptible to fluctuation as compared
to the non peak hours. If the charging of EVs is not properly optimized, the overall load will result into voltage drops and
ultimately in collapse of the whole distribution network.

The expected load due to the inclusion of EVs into smart grid can be minimized to great extent if EVs charging is properly

387
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

optimized. Optimally charged EVs not only enhance the smart grid power efficiency but can also provide other advance features like
ancillary services. EVs capability of vehicle-to-grid (V2G) capability (Su et al., 2012; Kempton and Tomić, 2005b) can contribute in
the local grid. In V2G capability, electric fleet can transfer and sell power to charging stations. But this demand a proper and more
sophisticated communications between EVs and charging station. Because not only the information regarding the current state of
energy level of EV is communicated but also the price of power and other availability information is exchanged between EVs and
charging stations.
EVs charging should also be properly scheduled and the whole charging and discharging process be optimally designed to avoid
any bottlenecks in distribution network. If the EVs charging is not properly optimized it will lead to severe harmonics, drastic load on
power grids, brownouts and blackouts, and drastic unwanted demand peaks (Hadley and Tsvetkova, 2009).

1.2. Contribution of this article

In this paper, we have provided a detailed discussion of various optimization approaches, objectives and approaches employed for
EVs charging. The main contributions of our paper can be summarized as follows:

• We discuss in detail EVs charging approaches (centralized/distributed). And the benefits of using these approaches have also been
incorporated.
• Various optimization approaches used during EVs charging have also been provided.
• We have also provided in depth discussion of the various optimization objectives achieved during the optimal charging of EVs.
The benefits of using various optimization approaches for EVs charging with various constraints considering different power levels
have been highlighted.
• Open issues, challenges, and future research directions related to EVs charging has also been highlighted.

Fig. 1 shows classification of our paper that includes EVs charging approaches and optimization objectives with optimization
approaches. Various acronyms used in the paper are also listed in Table 1. Table 2 shows the comparative view of this study with
existing surveys related to EVs charging. Authors in Rahman et al. (2016a) have explored the different optimization approaches
related to EVs charging. This study also encompasses a wide variety of optimization approaches applied to the EVs charging.
However, the different approaches related to EVs charging such as centralized and distributed EVs charging in conjunction with
optimization approaches have not been explored in much detail. Our study has provided a comprehensive overview of EVs charging
in comparison to this study. Authors in Tan et al. (2016), Wang et al. (2016) have discussed the different optimization objectives.
However, details of different optimizaiton objectives in conjunction with the EVs charging approaches (centralized, distributed, and
hybrid) have not been provided. Studies Habib et al. (2015), García-Villalobos et al. (2014) have highlighted the different EVs
charging approaches (centralized, distributed, and hybrid), while the optimization approaches with different optimization objectives
have not be discussed. Authors in Su et al. (2012) have also studied the optimization objectives without considering the optimization
objectives and optimization approaches. Our study addresses the limitations of all these studies and highlight the EVs charging with
the perspective of charging approaches, optimization objectives, and optimization approaches.

EVs Charging

EVs Charging Optimization Objectives Optimization Approaches

Approaches Achieved During EVs Charging For EVs Charging

Centralized Reduce Power Convex Optimiza-

EVs Charging Losses tion
Minimize Total
Cost of Energy
for Users Particle Swarm
Maximize Aggre-
Distributed gator Pro¿ts Optimization
EVs Charging Frequency
Regulation Genetic Algo-
Voltage Regulation rithms
Hybrid Support for
EVs Charging Renewable Energy Quadratic Pro-
Sources gramming
Avoid Distribution
Network Issues
Minimize the Peak Dynamic Program-
Load ming

Fig. 1. Organization of the paper with different optimization approaches and objectives used for EVs charging.

388
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Table 1
Abbreviations.
Symbol Description

ANN Artificial Neural Networks

BSS Battery Swapping Stations
CEAC Certainty Equivalent Adaptive Control
CFR Charging with Frequency Regulation
CPM Charging Post Manager
DSO Distribution System Operator
DR Demand Response
DSM Demand Side Management
DVC Decentralized V2G Capability
DGs Distributed Generators
DER Distributed Energy Resources
DLMP Distribution locational marginal pricing
EVs Electric Vehicles
EGVs Electric Ground Vehicles
FlC Fuzzy Logic Controller
GAs Genetic Algorithms
ICEV Internal Combustion Engine Vehicles
KKT Karush-Kuhn-Tucker
MPDIPA Modified Primal-dual Interior Point Algorithm
MPC Model Predictive Control
MILP Mixed Integer Linear Programming
MCOP Multi Constrained Optimal Path
MAS Multi-Agent System
MSS Maximum Sensitivities Selection
NHTS National Household Travel Survey
NNC Normalized Normal Constraint
OPF Optimal Power Flow
OLoCs Optimized Location Scheme for Electric Charging Stations
PEVs Plug-in Electric Vehicles
PSO Particle Swarm Optimization
PSO-CFA Particle Swarm Optimization with Constriction Factor Approach
PV Photovoltaics
RTSLM Real-Time Smart Load Management
SOC State of Charge
SQP Sequential Quadratic Programming
TCOPF Time Coordinated Optimal Power Flow
TOU Time of Use
V2G Vehicle-to-Grid
VFH Vector Field Histogram

Table 2
Comparison of existing surveys for EVs charging and optimization approaches for EVs charging.
Survey EVs charging Approaches Optimization Objectives Optimization approaches Publication year

García-Villalobos et al. (2014) 2014

Habib et al. (2015) 2015
Rahman et al. (2016a) 2016
Tan et al. (2016) 2016
Wang et al. (2016) 2016
Su et al. (2012) 2012

1.3. Organization of the paper

Rest of the paper is organized as follows: Section 2 discusses the various charging approaches used for EVs charging. Section 3
provides the overview of the various optimization objectives achieved and optimization methods used for achieving these objectives.
In Section 4, we have discussed in detail the optimization objectives with respect to the different optimization approaches, while
Section 5 highlights the various open issues, challenges, and future research direction related to EVs charging. Section 6 concludes the
whole paper.

2. EVs charging approaches

Optimized charging of EVs provides enormous benefits to the customer and also stabilizes the power grids. Reduction of load on
transformers, minimization of cost for energy, maximizing the user utilization, and reduction in the CO2 can only be achieved when

389
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

EVs are optimally charged (Richardson et al., 2012b).

To optimally charge EVs, a well defined EVs charging architecture has been integrated into the power grid. Aggregator unit (AU)
(Bessa and Matos, 2012; Guille and Gross, 2008) is now used as the central control device, which provides the facility of higher
number of EVs to connect with. AU provides the controlled smart charging capability to EVs. AU manages the optimal charging of
higher number of EVs, evenly distribute the load, and provides flexibility in cases the number of EVs increases. AU also provides the
ancillary services and contribute to the profit of the smart grid. Frequency regulation, voltage regulation, reducing power losses, load
levelling, and minimization of load variance are the well-known ancillary services provided by the AU. AU is regulated by another
controlling structure known as the distribution system operator (DSO). The main objective of DSO is to act as the central unit between
the transmission and the distribution power system (Lopes et al., 2011).
Taking into consideration the AU-based charging structure, the Optimal charging of EVs can be broadly classified into centralized
and distributed charging approaches (García-Villalobos et al., 2014). The main difference between the centralized and distributed
approach is in the aggregator function and the location of the EVs charging decision.

2.1. Centralized EVs charging approaches

Centralized EVs charging approach discussed in Hutson et al. (2008), Ko and Keun Sung (2013), Sundstrom and Binding (2010),
Acha et al. (2010), Nguyen and Le (2014), Han et al. (2010a), Sortomme and El-Sharkawi (2011), Wu et al. (2012b), Tushar et al.
(2014), Huang et al. (2012), Galus and Andersson (2008) provides the centralized control and management capability to the AU. In
this approach, AU collects all the charging related information of EVs. SOC of batteries, location of EVs, their parking time, and
estimation of number of EVs related information is collected by the AU with help of charging post manager (CPM). CPM facilitates the
AU in controlling the number of different cars in the parking area.
A safe estimation for electricity demand is performed by AU in centralized approach. Hence, AU also takes part in management of
power grid. Forecast for power demand is made by optimization approaches employed by centralized methods. This forecasts de-
manded by AU is then transformed to the DSO, which informs the other units of power structure for smooth charging of expected EVs
in parking area. AU in centralized approach, also takes part in ensuring the smooth operation of distribution system. In case of any
fault in distribution system, AU then halts all the charging sessions and takes part in the actions to remove faults in the distribution
system.
Different optimization algorithms in centralized approach have been used to achieve different objectives. The main objectives of
major optimization approaches in centralized EVs charging method are the enhancement in AU profit, minimization in costs, and
minimization of load on the distribution system.
The provision of ancillary services such as frequency control, voltage control, and battery states are well supported by centralized
approach. By this feature, the AU maximizes its profit. Another advantage of using the centralized EV charging approach is the
maximum usage of available network capacity.
There are also certain limitations of centralized EVs charging approach. If AU fails, then the whole structure suffers. Therefore,
alternative arrangements are necessary, that in turn increases the cost of the system. AU has to process an enormous amount of
information, so it demands a well sophisticated communications infrastructure, which adds complexity in the system. When the
number of EVs increases, it adds more complexity in the centralized architecture. For this purpose, enormous processing has to carry
out by AUs and other components. There are also the security and privacy related issues with the centralized EVs charging approach.

2.2. Distributed EVs charging approaches

In distributed or decentralized EVs charging approach (Gan et al., 2011; Gerding et al., 2011; Jiang and Fei, 2013; Ardakanian
et al., 2013; Ma et al., 2010; He et al., 2012; Hu et al., 2014; Yang et al., 2014; Tan et al., 2014; Qi et al., 2014; Rezaei et al., 2014;
Wen et al., 2012) EV itself, not the AU makes all the decisions regarding the charging or discharging. Therefore, it demands some type
of computational intelligence on the part of EV or customer. It addresses the data security issues, that the centralized approach faces
in charging the EVs.
Distributed approach for EVs charging adds more flexibility and scalability in the system. As the number of cars increases, this
approach distributes the computational load on the customer which minimizes the overall processing and computational load on the
charging network. Robustness in the distribution network also enhances many fold when EVs are charged in the distributed archi-
tecture. This approach makes the whole charging operation more fault tolerance compared to the centralized approach.
The whole charging infrastructure for this approach is not very complex. A simple and precise communications and distribution
structure fulfils the required demand of optimized EVs charging. Main charging decisions are taken solely by the customer itself. To
what extent and to when is charged, is decided by the customer. Due to these advantages, distributed EVs charging approach have
gained much attention from users compared to the centralized approach.
Distributed approach has also certain limitations. The ancillary services are not very well supported in this approach. Only some
services are provided but in limitations. The main limitation of this approach is the avalanche effect that minimizes the system
performance. Avalanche effect is the increase in the load, when the large number of cars are charged at the low electricity rates.

2.3. Hybrid EVs charging

Hybrid approaches for EVs charging have also been introduced to optimally charge the EVs. Hybrid approaches employ both

390
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Table 3
Centralized EVs charging approach with its different capabilities.
References V2G Main objective achieved Time scheduled Local Generation Publication year

Hutson et al. (2008) Yes Maximize aggregator profits Real-time No 2008

Ko and Keun Sung (2013) No Achieving fairness in EVs owners and minimizing total cost Real-Time No 2013
of energy for users
Sundstrom and Binding (2010) No Minimizing total cost of energy for users and frequency Day-ahead Yes 2010
regulation
Acha et al. (2010) Yes Reducing power losses and avoiding distribution network Day-ahead No 2010
issues
Nguyen and Le (2014) Yes Minimizing the total cost of energy for users Day-ahead No 2014
Han et al. (2010a) Yes Maximizing aggregator profits Real-time No 2010
Sortomme and El-Sharkawi Yes Maximizing aggregator profits and voltage regulation Real-time Yes 2011
(2011)
Wu et al. (2012b) No Minimizing total cost of energy for users Day-ahead No 2012
Tushar et al. (2014) Yes Minimizing total cost of energy for users and avoiding Day-ahead Yes 2014
distribution network issues
Huang et al. (2012) No Minimize generation cost Real-time No 2012
Galus and Andersson (2008) No Minimize total cost of energy for users Real-time Yes 2008
Gonzalez Vaya and Andersson No Minimizing the generation cost and regulating the voltage Real-time No 2012
(2012)
Nguyen and Bin (2012) Yes Minimizing the peak load and avoiding distribution network Day-ahead Yes 2012
issues
Soares et al. (2014) No Frequency regulation Real-time No 2014
Kang et al. (2013) Yes Minimizing the peak load and avoiding distribution network Real-time Yes 2012
issues
Richardson et al. (2012a) No Avoid distribution network issues and Lowering CO2 Day-ahead No 2012
emissions
Le et al. (2015) Yes Minimizing the peak load and Supporting renewable energy Real-time No 2015
sources for charging of EVs
Huang et al. (2014b) No Avoiding distribution network issues and congestion Day-ahead No 2014
management

centralized and distributed approach for the charging of EVs. The main advantage of hybrid EVs charging is that, it addresses the
various limitations of centralized and distributed EVs charging approach.
In centralized EVs charging, all the decisions are made by the central unit. Therefore, the central unit has to be aware about the
requirements of all the customers. Central unit has to gather all the information to take the charging decisions. This helps in optimal
charging of EVs. However, the main drawback of this approach is that, if the central unit encounters some problem, then the whole
system suffers and witness the blockage. This limitation has now been addressed with the introduction of hybrid EVs charging. Hybrid
EVs charging has been discussed in Awasthi et al. (2016). In this scheme, the genetic algorithm and particle swarm optimization has
been jointly considered to solve the optimization problem for optimally charging the EVs.
Hybrid EVs charging approaches also addresses the limitations of distributed EVs charging methods. The provision of ancillary
services such as voltage and frequency regulations have also been discussed with respect to hybrid distributed EVs charging ap-
proaches. Avalanche effect, that minimizes the performance of the system can also be addressed with the help of hybrid EVs charging
approaches.

3. Overview of optimization objectives and optimization methods applied for EVs charging

3.1. optimization objectives

With the centralized and distributed EVs charging approach, various optimization approaches are used in either centralized or
distributed fashion for EVs charging. These optimization approaches are employed to gain one or more specific objective during
charging process. To enhance the efficiency of EVs, power grid, and aggregator, efficient optimization approaches are usually used.
These optimization approaches are supported by the centralized or distributed EVs charging process. Table 3 encompasses the
centralized EVs charging approach with its main contribution in terms of achieving various optimization objectives, while Table 4
encompasses the distributed EVs charging approach with its main contribution in terms of achieving various optimization objectives.
These charging approaches have also been explored in context to their support for the local generation capability. EVs can also be
charged with the help of some local supported renewable energy sources such as wind and solar energy sources (Kaschub et al.,
2016). These renewable energy sources have been studied under the local generation.
An extensive work has been done on the EVs charging. The state-of-art work involving the EVs charging also considers the
different optimization techniques involved for EVs charging. Current work on EVs charging involving integration and allocation of
renewable-energy sources has been explored in Mozafar et al. (2017). Other dimensions of EV charging such as forecasting charging
demand (Hadi Amini et al., 2016), particle-swarm optimization-based charging pattern (Rahman et al., 2016b; Rahman et al., 2015),
hybrid charging (Awasthi et al., 2016), optimal power allocation for EVs charging (Vasant et al., 2016), wireless power transfer for

391
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Table 4
Distributed EVs charging approach with its different capabilities.
References V2G Main objective achieved Time scheduled Local Generation Publication year

Gan et al. (2011) No Minimizing the peak load Day-ahead No 2011

Gerding et al. (2011) No Maximizing the aggregator profits Real-time No 2011
Jiang and Fei (2013) No Minimizing total cost of energy for users and reducing power Day-ahead Yes 2013
losses
Ardakanian et al. (2013) No Maximize aggregator profits Real-time No 2013
Ma et al. (2010) No Minimize the peak load and voltage regulation Day-ahead No 2010
He et al. (2012) Yes Minimizing total cost of energy for users and reducing power Day-ahead No 2012
losses
Hu et al. (2014) No Minimizing total cost of energy for users Day-ahead No 2014
Yang et al. (2014) No Minimizing the peak load and minimizing total cost of energy Day-ahead No 2014
for users
Tan et al. (2014) Yes Supporting for Renewable energy sources and minimizing Real-time Yes 2014
total cost of energy for users
Qi et al. (2014) No Maximize aggregator profits and avoiding distribution Real-time No 2014
network issues
Rezaei et al. (2014) No Minimize total cost of energy for users Real-time No 2014
Wen et al. (2012) No Minimize generation cost Real-time No 2012
Gan et al. (2012) No Minimizing the peak load and total cost of energy for users Day-ahead No 2012
Zhou and Cai (2014) No Regulating voltage and avoiding distribution network issues Real-time No 2014
Beaude et al. (2013) Yes Regulating voltage and avoiding distribution network issues Real-time Yes 2013
Tokudome et al. (2009) Yes Regulating the frequency and voltage regulation Real-time Yes 2009
Liu et al. (2013a) Yes Regulating the frequency and minimizing the peak load Day-ahead No 2013
Luo et al. (2014a) Yes Regulating the frequency and integration of renewable Real-time No 2014
energy sources
Wu et al. (2012a) Yes Regulating the frequency and minimizing the total cost of Real-time No 2012
energy for users
Mullen (2009) Yes Regulating the frequency Real-time Yes 2009
Lin et al. (2014) Yes Regulating the frequency and avoiding distribution network Real-time No 2014fa
issues
Ahn et al. (2011) No Regulating the voltage and minimizing the peak load Day-ahead No 2011
Karfopoulos et al. (2011) No Regulating the voltage and minimizing the peak load Day-ahead and real- Yes 2011
time
Fan (2012) No Supporting renewable energy sources and minimizing the Real-time No 2012
peak load
Iversen et al. (2014) No Avoiding distribution network issues Day-ahead No 2014
Torabi and Gomes No Minimizing the peak load and generation cost Real-time Yes 2014
(2014)
Mierau et al. (2012) Yes Minimizing the peak load Real-time No 2012

EVs charging (Lu et al., 2018), usage of dynamic charging coils (Zhang et al., 2017), multi-server fair queuing based charging (Wang
et al., 2017), crowd sensing-based charging (Yang et al., 2017), distributed charging on highways (Gusrialdi et al., 2017), charging
involving publish or subscribal communication networks (Cao et al., 2015), distributed online learning (Ma et al., 2016), improving
EVs charging considering energy storage (Sabillon et al., 2017), capacity sizing for optimal charging (Ugirumurera and Haas, 2017),
novel bidirectional EVs charging (Singh, 2016), probabilistic charging (Arias et al., 2017), routing impact on EVs charging (Conway,
2017), heterogeneous EVs charging (Malhotra et al., 2017), privacy preservation during EVs charging (Li et al., 2017), and two-stage
optimal stochastic EVs charging (Wu and Sioshansi, 2017).
Different objectives for optimal charging of EVs with their optimization approaches are given as follows.

3.1.1. Reduce power losses

Optimal charging of EVs can handle the power losses that happens during the charging of EVs. Power losses handling is discussed
in Acha et al. (2010), Oliveira et al. (2013), Deilami et al. (2011), Sortomme et al. (2011), Clement-Nyns et al. (2010), Grau Unda
et al. (2014), Singh et al. (2010), Clement et al. (2009b), Liu et al. (2013b), Moses et al. (2010), Zhan et al. (2012), Dixit and Roy
(2015), Bashash and Fathy (2013), Pourabdollah et al. (2013), Murgovski et al. (2013), Ko and Jang (2013), Li et al. (2012), Zhu et al.
(2004), Brembeck and Ritzer (2012), Neaimeh et al. (2013). Different optimization approaches are used during the charging of EVs to
minimize the power losses. Optimized charging of EVs not only reduces the power losses of EVs battery but can also improve the
quality of distributing network. The objective of reducing the power losses during optimized EVs charging also increases the effi-
ciency of the distribution network. The power losses in EVs charging due to optimization approaches are 3 times lower than the ICEV.
In EVs, the energy is not consumed while the vehicle is idle, however in ICEV various energy losses occur even when the vehicle is
stationary. Various optimization approaches are taken into consideration to conserve the energy in EVs when it is on the road. EVs
batteries which are usually secondary (rechargeable) batteries account for various power losses in EVs. Various optimization ap-
proaches have also been defined in this paper that optimize the power losses in EVs batteries.

392
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

3.1.2. Minimize total cost of energy for users

Optimization methods for EVs charging for energy cost minimization are discussed in Sundstrom and Binding (2010), Nguyen and
Le (2014), Wu et al. (2012b), Tushar et al. (2014), Jiang and Fei (2013), He et al. (2012), Hu et al. (2014), Yang et al. (2014), Tan
et al. (2014), Rezaei et al. (2014), Bessa et al. (2011), Di Giorgio et al. (2014), Gonzalez Vaya and Andersson (2012), Gan et al.
(2012), Sheikhi et al. (2013), Dallinger and Wietschel (2012), Zhou and Cai (2014), Beaude et al. (2013), Nguyen and Bin (2012),
Bashash and Fathy (2014), Wei et al. (2016), Siddiqi et al. (2011a), Wu et al. (2015), Siddiqi et al. (2011b). Charging cost plays an
important role in selection of charging station as well as in selection of EVs. Using EVs is cost efficient as compared to ICEV, but the
charging cost of EVs during the peak or off-peak hours varies. Different optimization approaches for charging EVs at reasonable cost
in peak and off-peak hours have been taken into account to give the optimal solution to customers. Optimization approaches for EV
charging at different locations regarding the price of charging have also been presented. Some customers usually charge their cars at
various locations for example at home and at the parking lots at various charging cost. The cost for charging EVs by charging stations
is defined by various rates. Some defined the charging cost by the hour, some used the kilowatt-hour to set the price for charging EVs,
others use the session fee, subscription for month and year fee. And there are also the charging stations which are quite free.

3.1.3. Maximize aggregator profits

Optimization approaches for enhancing the profits of charging stations or aggregators during EVs charging have been presented in
Hutson et al. (2008), Han et al. (2010a), Sortomme and El-Sharkawi (2011), Qi et al. (2014), Han et al. (2010b), Sortomme and El-
Sharkawi (2012a,b), Soares et al. (2014), Jin et al. (2013), Vagropoulos and Bakirtzis (2013), Bessa et al. (2012), Bessa and Matos
(2013), Balram et al. (2013), Deshpande and Murali (2013). Aggregators increase their profits by optimally charging the large
number of EVs. Different ancillary services provided by aggregator during optimized charging of EVs also contribute to the profits of
charging units. By adopting the efficient optimization methods, aggregators not only maximize their profits but also overhaul its
structure for maintenance purposes. The charging stations not only enhances their profits by optimally charging EVs but also con-
tribute to limit the impact on power grids. The optimization approaches for V2G capability in conjunction with the aggregator and
power grid have been reviewed with context to maximize the profit for aggregator.

3.1.4. Frequency regulation

EVs can contribute to the frequency regulation of the power grid (Han et al., 2010b; Sortomme and El-Sharkawi, 2011; Kang et al.,
2013; McArthur et al., 2007; Lopes et al., 2009; Tokudome et al., 2009; Pillai and Bak-Jensen, 2010a; Liu et al., 2013a; Luo et al.,
2014a; Wu et al., 2012a; Rotering and Ilic, 2011; Donadee and Ilić, 2012; Pillai and Bak-Jensen, 2010b; Jin et al., 2013a; Verzijlbergh
et al., 2012; Mullen, 2009; Celli et al., 2012; Lin et al., 2014; Mojdehi et al., 2015; Han et al., 2011). Frequency is the number of times
in seconds that the alternating current takes in its flow in the distribution network. For smooth operation, a 60 Hz of constant
frequency is required in power grid. Optimization approaches are required in balancing the generation and load demand. For this
purpose, a frequency regulation approach is needed. Distribution networks encounter certain frequency imbalances, due to the huge
differences in generation and consumption. EVs mostly with the help of V2G capability help the power grid in regulating the
frequency. Various optimized approaches have been surveyed that provide the frequency regulation during EVs charging or dis-
charging mode. By frequency regulation of power grid the electricity rates for other users can be hold down. Provision of frequency
regulation by EVs has also certain hurdles.

3.1.5. Voltage regulation

Optimal charging of EVs can also provide the voltage regulation to residential and other power grids. Voltage regulation achieved
during the optimized EVs charging approaches has been discussed in Richardson et al. (2012b), Beaude et al. (2013), Singh et al.
(2012), Ahn et al. (2011), Karfopoulos et al. (2011), Abe and Saitoh (2014), Huang et al. (2013), Mitsukuri et al. (2012), Cao et al.
(2012), Vagropoulos et al. (2016), Jiménez and García (2012), Cheng et al. (2015), Fazeli et al. (2015), Herrera et al. (2015), Bai and
Qiao (2015), Saberbari and Saboori (2014). Various utilities in the power grid require a constant voltage balance for their smooth
operation. The integration of large number of EVs in power grid can have significant impact on the voltage balance of overall grid.
However, EVs can also contribute in regulation of voltage. With the help of V2G capability, EVs use various optimization algorithms
in regulating the voltage of the power grid. The voltage drop during the peak hours is taken into consideration by various optimi-
zation approaches, and then EVs are optimally recharged. The optimal charging of EVs not only regulate the voltage rise and drop but
also minimizes the impact of EVs on the power grid.

3.1.6. Support for renewable energy sources for charging of EVs

To limit the cost and emissions, EVs can be recharged from renewable energy based charging units. Optimization approaches
supporting the renewable energy sources for EVs charging are discussed in Lopes et al. (2009), Luo et al. (2014a), Fan (2012), Huang
et al. (2014a), Jin et al. (2014, 2013b, 2012), Zhang et al. (2014a), Lu et al. (2014), Arikan et al. (2015), Liu et al. (2014), Nguyen
et al. (2015), Richardson (2013), Kempton and Tomić (2005a), Mauri and Valsecchi (2012), Costa et al. (2014), Huang et al. (2015),
Yan et al. (2013), Zhao et al. (2015), Ratnam et al. (2013). EVs have a significant impact on the power grid. The traditional power
grid have limited power supply. Therefore, to avoid the voltage and frequency drop during the peak hours and to reduce the carbon
emission in the environment, various renewable energy sources such as light, wind, heat, and other sources are used to charge EVs. To
optimally exploit these renewable energy sources for EVs charging, various optimization approaches have been employed to enhance
the efficiency in the system. EVs that themselves by using the renewable energy sources on board or at home, can also contribute to
the power grid. In this way, EVs can take part in stabilizing the power grid.

393
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

3.1.7. Avoid distribution network issues

Charging of EVs during the peak hours can lead to a significant impact on the power grid working. But the optimal charging of EVs
can contribute in reducing the various issues relating to power grid degrading inefficiency. Optimization approaches for minimizing
the distribution network issues are discussed in Oliveira et al. (2013), Deilami et al. (2011), Singh et al. (2010), Zhou and Cai (2014),
Soares et al. (2014), Lopes et al. (2011), Richardson et al. (2012a), Iversen et al. (2014), Lopes et al. (2007), Sundstrom and Binding
(2012), Mets et al. (2010), Hu et al. (2012), Zhang et al. (2012), Qian et al. (2011), Tian et al. (2012), Freire et al. (2010), Shaoyun
and Liu (2012), Bashash et al. (2011), Rahimi et al. (2014), Ruiqin and Zhongjing (2015), Mehar and Senouci (2013), Pourazarm
et al. (2014), Sun et al. (2014), Huang et al. (2014b), Cai et al. (2013), Luo et al. (2014b), Dutta and Boulanger (2014). Distributing
power network issues such as irregular power flow, voltage and power lines fluctuation, defected feeders and buses, bottleneck in the
communication structure can be optimally avoided with optimized EVs charging. By minimizing various issues, not only the effi-
ciency of the power grid is increased, but also the cost of energy for EVs charging and other users also decreased.

3.1.8. Minimize the peak load

Optimization approaches for EVs charging to minimize the load on power grid are discussed in Gan et al. (2011), Ma et al. (2010),
Mets et al. (2010), Torabi and Gomes (2014), Le et al. (2015), Yuqing et al. (2014), Mierau et al. (2012), Kazerooni and Kar (2012),
Chan et al. (2015), Zhan et al. (2015), Chen et al. (2012), Su et al. (2013), Said et al. (2014), Chammas et al. (2013), Sojoudi and Low
(2011), Mehboob et al. (2014), Feng et al. (2013), Mets et al. (2012), Muñoz and Ruspini (2012), Nguyen et al. (2014). Charging of
large number of EVs at the peak time can have a significant impact on the power grid. This can lead to voltage and frequency drops
and can also decrease the efficiency of power grid. Overloading can disturb the various devices in power grid. The impact on
transformers due to overloading can interrupt the power supply to other commercial and home users. Therefore, different optimi-
zation approaches have been incorporated into EVs charging to minimize the load on the power grid. The V2G capability of EVs also
comes out to be very helpful in reducing the peak load on the power grid. With the help of V2G capability, EVs can contribute their
power to the power grid during the peak hours. Optimization approaches with V2G capability in EVs have also been discussed with
their new trends. Fig. 2 highlights various optimization objectives achieved during EVs charging.

3.2. Optimization methods applied to EV charging

Optimization can be defined as, “mathematically finding the inputs of the fluctuating functions (function whose value is subject to
maximize or minimize) that are under various constraints (Pardalos and Resende, 2002)”. As its name suggests, but it does not mean
that optimization is the finding of the best possible or optimum solutions for a given complex problem. As there are certain problems
whose solutions are not possible. Those problems fall into the category of the NP-hard problems (Garey and Johnson, 1979). For these
NP-hard problems, no polynomial time algorithm exists. These problems need the exponential time which is very high, for their
solution. To address this issue, various approximate methods such as heuristic methods and artificial neural networks (ANN) have
been proposed in the literature (Pearl, 1984). As compared to conventional optimization approaches like linear programming and
quadratic programming, the heuristic methods are considered to be the simple and satisfactory solutions for various complex pro-
blems. Meta-heuristics are the more general form of heuristic methods. Meta-heuristics are used to find the optimal solutions with
limited modifications in the original problem. However, sometimes the heuristic and meta-heuristics methods do not provide the
simple and satisfactory solutions for various complex problems. For these type of problems, the parallel processing is then employed
to achieve the better solutions.

Fig. 2. Optimization objectives achieved by applying the various optimization approaches.

394
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Computational optimization have gained much attention in todays research domain due the advancement in the computer world
and in the algorithms. Computational optimization is the set of procedure that consists of designing, implementation, and then
evaluating the problem (Banos et al., 2011).
EVs charging is now considered as the optimization problem and is solved by various optimization approaches. Various opti-
mization approaches take into consideration various objectives during optimization problem of EVs charging. In many cases, two or
more optimization approaches provide the solution for a single objective. While in many other EVs charging optimization problem, a
single optimization approach provide the solution for multiple objectives. A brief overview of various optimization approaches used
for EVs charging with their references are given as follows:

3.2.1. Convex optimization

EVs charging through convex optimization is discussed in Wen et al. (2012), Sortomme et al. (2011), Bessa et al. (2011), Gonzalez
Vaya and Andersson (2012), Gan et al. (2012), Sortomme and El-Sharkawi (2011), Soares et al. (2014), Ahn et al. (2011), Richardson
et al. (2012a), Zhang et al. (2014b), Sojoudi and Low (2011), Wei et al. (2016), Pourabdollah et al. (2013), Murgovski et al. (2013).
Through convex optimization in EVs charging various objectives have been achieved and has resulted into the enhancement in the
system efficiency. Convex optimization is much similar in its problems solving method to the least square or linear programming. If
the problem is formulated as the convex optimization problem, then it can be solved easily for finding its solutions. Therefore, the
given problem after making it convex can be solved just like the least square problems. However, the main limitation of this approach
is difficulty in formulating the problem as the convex optimization problem. Once, the problem is convexed, then it can be solved
using the least square or linear programming.

3.2.2. Meta-heuristic method

The meta-heuristic optimization approach for EVs charging is studied in Valentine et al. (2011). Meta-heuristic methods are used
to solve those optimization problems that have incomplete information. These approaches are called as the higher-level methods, that
can be implemented for finding the optimal solution. Larger set of solutions are sampled and then some assumptions are made
regarding these solutions. Some of the global search meta-heuristic optimization approaches for achieving the objective of optimal
EVs charging are given below:

• Particle swarm optimization: Particle swarm optimization (PSO) for EVs charging is discussed in Grau Unda et al. (2014), Celli
et al. (2012), Tian et al. (2012), Dixit and Roy (2015), An et al. (2013), Ko and Jang (2013), Siddiqi et al. (2011a), Wu et al.
(2008). PSO is the stochastic method-based optimization approach that is based on the social behavior of fish schooling and birds
flocking. In PSO, the optimal solution is iteratively improved with respect to the optimal problem quality. The optimal solutions
are named as the particles are placed in specific position in the search space. The movement of these particles are then judged with
respect to other particles. The best known position is then assigned to the particles. In this way, the swarm comes out with the
optimal solution for a specific optimized problem.
• Genetic algorithms: Genetic algorithms (GAs) for optimized EVs charging are discussed in Gonzalez Vaya and Andersson (2012),
Mehar and Senouci (2013), Jain et al. (2009), Desai et al. (2010), Wu et al. (2015), Siddiqi et al. (2011b), Mehboob et al. (2014),
Yan et al. (2013), Herrera et al. (2015), Huang et al. (2006). Base on the natural selection and genetics, GAs use the random search
in solving the optimized problem. This random search is not a purely randomized search in the search space, but GAs use the
historical data in searching the better solution in the search space. The GAs are considered to be the most robust compared to
other searching algorithms used for finding the solution of an optimized problems.
• Artificial immune system: The optimized EVs charging via artificial immune system is discussed in Oliveira et al. (2013). In this
optimized EVs charging approach, the immune system capability of memory and learning is used in problem solving.

3.2.3. Dynamic programming

EVs can also be optimally charged using the dynamic programming (Clement-Nyns et al., 2010; Han et al., 2010b; Beaude et al.,
2013; Pourazarm et al., 2014; Sun et al., 2014; Li et al., 2012; Sweda and Klabjan, 2012; Feng et al., 2013). In dynamic programming,
the optimization problem is divided into multiple sub-problems, and then the sub-problems are solved and their solutions are stored
in the memory. These solutions of the sub-problems are somehow indexed in the memory, and are used in the future for these kind of
sub-problems. Using dynamic programming, usually the computation time can be minimized for solving a particular problem. Op-
timization problems that have the features of overlapping sub-problems and optimal substructures can be solved using the dynamic
programming.

3.2.4. Quadratic programming

In Clement-Nyns et al. (2010), Sundstrom and Binding (2012), Mao et al. (2014), Bashash and Fathy (2013), Bashash and Fathy
(2014), Han et al. (2011), Zhao et al. (2015), Ratnam et al. (2013), Zhu et al. (2004), Mayr et al. (2011), Huang et al. (2014b),
Brembeck and Ritzer (2012), Bai and Qiao (2015), Deshpande and Murali (2013), Mets et al. (2012), Cai et al. (2013), Saberbari and
Saboori (2014), EVs are usually optimally charged through the quadratic programming. In quadratic programming, the quadratic
function is usually optimized by considering the linear constraints on its variables. Quadratic problems can be solved on their own, or
can be optimally solved by dividing them into sub-problems by considering the general constraints.

395
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

3.2.5. Fuzzy logic

Fuzzy logic for solving the optimal charging of EVs problem is discussed in Zakariazadeh et al. (2014), Singh et al. (2012), Luo
et al. (2014b), Lee and Sul (1998), Jiang et al. (2010), Muñoz and Ruspini (2012). Fuzzy logic is used to give the degree of partial
truth compared to complete true or false, as is the case in boolean logic. Fuzzy logic is the combination of other many valued logics,
where the values between the 0 and 1 are used to test the degree of truth logic in solving the problem. Usually, the fuzzy logic are
considered as the infinite-valued logic for finding the optimal solution of optimization problem.

3.2.6. Heuristic method

EVs can be charged optimally using the heuristic methods. Various heuristic methods for EVs charging are discussed in Acha et al.
(2010), Deilami et al. (2011), Singh et al. (2010), Clement et al. (2009b), Liu et al. (2013b), Soares et al. (2014), Singh et al. (2012),
Kang et al. (2013), Lopes et al. (2011), Subramanian et al. (2012). Heuristics can be employed to gain the immediate solutions for
optimization problem. The solution provided by the heuristic method may or may not be the optimal solution. When it seems that, the
solution for the one specific problem is almost impossible, then some heuristics are employed to find out the practically immediate
solution for the problem.

3.2.7. Game theory

Optimized EVs charging with the implementation of game theory is discussed in Grau Unda et al. (2014), Sheikhi et al. (2013),
Zhou and Cai (2014), Nguyen and Bin (2012), Karfopoulos et al. (2011), Dutta and Boulanger (2014). Game theory is the logic that
base on the conflict and cooperation between the participants. It is not only applied to computer science, but also is used in the
economics, political science, biology, and psychology. In the beginning, game theory addresses only the gains of one person, that
results in the complete loss of their participants. This was named as the zero-sum games. Game theory in most of its applications deals
with the interaction of group of people. With respect to interaction, two branches of game theory, the cooperative and non-co-
operative game theory are discussed in the literature.

3.2.8. Graph search algorithm

There are other optimization approaches that can be employed for achieving the optimized EVs charging. For example, graphs
search optimization approach is also employed in EVs charging (Luo et al., 2014a; Neaimeh et al., 2013). During this optimization
approach, different objectives are achieved that provide help in stabilizing the power grid.

4. Objective-based optimization of EVs charging

4.1. Optimization approaches used for optimizing the power losses

Various optimization approaches have been discussed in the literature to minimize the power losses during EV charging. The
controlled power flow provided by the optimized EV charging overcomes the power losses during the charging of EVs from power

Optimization Techniques used for EVs recharging . Such

as Convex optimization , Dynamic optimization , Particle
swarm optimization and etc .

(i)

(ii)

(iii)

Recharging Points used for Evs Recharging

(iv) Different Energy Sources used for EVs recharging such as

(i) Wind (ii) Solar (iii) Power Plant and (iv) Storage batteries

Fig. 3. Optimized EVs charging using the conventional and renewable energy sources.

396
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

grid. The optimized EV charging not only minimizes the power losses but also enhances the efficiency of the power distribution
system. An optimized EVs charging process has been shown in Fig. 3. The optimization approaches employed to minimize the power
losses during the EV charging are discussed as follows:
Time coordinated optimal power flow (TCOPF) (Acha et al., 2010) is used to control EVs batteries and charging infrastructure.
The energy storage resources which are at the longer distance from the charging points are optimally coordinated. In TCOPF, EVs are
optimally charged by the distribution network in a specific time interval to minimize the peak load and power loss. The TCOPF tool
can also be utilized by the distribution network operators to estimate the impact of EVs on the power distribution system. A real-time
smart load management (RT-SLM) (Deilami et al., 2011) approach is introduced to minimize the power loss and potential load on the
smart grid. The integration of multiple EVs into smart grid can cause a potential thread to the security and performance of smart grid.
In this optimized EV charging, an optimal network criteria is selected to reduce the impact of large number of EVs charging from the
smart grid. EVs satisfying the network criteria are only charged for some specific time interval. Authors in Singh et al. (2010), have
compared the impact of optimized and coordinated EVs charging with the un-coordinated EVs charging. The various numerical
results show that the optimal charging not only reduces the power losses of distribution power systems but also controls the voltage
and frequency fluctuations. In Clement et al. (2009b), stochastic programming is used to achieve the optimal charging of EVs in
residential distribution grid. EVs which are charged from household charging units, usually arbitrary choose the charging periods
using this approach. Through this optimal charging, not only the power losses are controlled but also the local grid load factor has
improved. By using the stochastic programming, the expected daily load is forecasted and arbitrary periods are assigned that results
into the reduction of power losses. A modified primal-dual interior point algorithm (MPDIPA) is used in Liu et al. (2013b) to optimal
plan the charging stations locations for EVs. The sites for the charging stations for EVs are optimally selected. This optimal planning
of sites for EVs charging stations minimizes the power losses that can result from the transmission of power to the distant charging
stations.
Artificial immune systems are also used to reduce the power losses during EVs charging (Oliveira et al., 2013). In this method, the
load is distributed over various charging periods. The installation of capacitors and the optimal flow of power for charging of EVs
using the artificial immune systems not only reduces the power losses but also stabilizes the voltage of the smart grid. In Grau Unda
et al. (2014), the reduction in the power losses is achieved through the PSO. An agent-based approach using the PSO, defines the
charging schedule of EVs batteries. The batteries of EVs are charged in an optimized way, this not only reduces the power losses but
also conserves the energy and minimizes the total cost of energy for users. This optimal multi-agent design for EVs charging is tested
in a test bed of laboratory micro-grid. The particle swarm optimization with constriction factor approach (PSO-CFA) (Dixit and Roy,
2015) discusses the optimal charging and discharging of EVs. EVs are placed in the distribution network in an optimal way to
minimizes the power losses and to stabilize the frequency and voltage of the power grid. PSO-CFA based placement of EVs on IEEE-
333 radial distribution test system shows the reduction in the power losses compared to other charging approaches. EVs can also be
optimally charged through wireless power transmission using the PSO (Ko and Jang, 2013). The online EVs have the potential to
charge themselves while in motion. These EVs are now finding their way in the mass transportation system (Jamil et al., 2017). The
wireless power transfer is more susceptible to the power losses compared to other approaches for EVs charging. However, this
optimization problem is solved through the PSO.
Convex optimization approach is also used to reduce the power losses in the power distribution system. In Sortomme et al. (2011),
the authors have used the convex optimization to reduce the power losses during EVs charging. The three optimal charging algo-
rithms are used to coordinate the charging of EVs. The relationship between the distribution system losses, expected load, and voltage
variance is studied with reference to the optimal charging of EVs. In Zhan et al. (2012), authors have presented the convex quadratic
programming to optimize EVs charging. The effect of optimized EVs on the economic gains of the power grid have been study from
various perspectives. In this case, the IEEE 33-bus test case has been implemented and the nodal voltage is corrected iteratively using
the convex optimization approach. This optimized EVs charging approach, not only minimizes the power losses, but also lower the
load variance. The various energy losses of the powertrain can also be minimized with the help of convex optimization (Pourabdollah
et al., 2013). The losses by the electric motors and combustion engine are optimized in relation to the sizes of the various components
of powertrain. A convex optimization-based powertrain model is used to estimate the quadratic losses of the various components of
the powertrain. An optimized driving pattern is also suggested using this model based on the charging state of the powertrain. To
conserve the energy in the powertrain a convex optimization-based pontryagin’s maximum principle is used to solve the optimization
problem of optimal charging and minimization of power losses (Murgovski et al., 2013). The integer variable that is directly con-
trolled is the engine on/off state for the optimal power split control. The optimization approach is applied on the four bus lines of
powertrain to estimate effect of optimal charging and the power losses. The relation between the power split control and the battery
size is also simulated to determine the effect of optimal engine on/off state in the reduction of power losses.
Dynamic programming is used in Clement-Nyns et al. (2010) to analyze the impact of optimal EVs charging on the residential
distribution grid. Non-coordinated charging of EVs can lead to deviations and power losses. In this proposed approach not only the
power losses are minimize but the load factor of the main grid is also maximized using the dynamic programming. For optimal
forecasting of the domestic load, the stochastic programming is employed. Electric ground vehicles (EGVs) can be optimally charged
while considering the dynamic programming (Li et al., 2012). The optimal usage of battery power is taken into consideration to
maximize the travel distance to reach a destination (charging station). The power losses of these in-wheel motors are reduced
significantly by optimally distribution of actuation torque. The actuation torque and varied velocity of these in-wheel motors are
optimally solved using the dynamic programming to conserve the energy by minimizing the power losses.
Demand response (DR) optimization for EVs charging can be achieved using the quadratic programming (Bashash and Fathy,
2013). In this optimization approach the power split model is used to reduce the power losses especially the Ohmic energy losses in

397
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

EVs batteries. The optimization approach is applied on the national household travel survey (NHTS) data to optimally design EVs
charging and mobility pattern that consumes less energy and minimize the power losses. The sequential quadratic programming
(SQP) is used in Zhu et al. (2004) to optimize EVs charging. The optimal power distribution in EVs using the SQP maximize the load
factor, reduces energy cost, and minimizes the power losses in the system. The four-step method based on the quadratic problem is
used in this method to optimally manage the energy in EVs and in power train hybrid vehicle. The fuzzy rule system is achieved from
the numerical results. The energy losses in robotic EVs can also be minimized with the help of nonlinear control allocation (CA)
algorithm (Brembeck and Ritzer, 2012). Using the quadratic programming this CA algorithm optimally control the energy flow in the
robotic EVs. The power losses are usually minimized by estimating the average energy consumption along the pre-calculated path and
defining the charging cycles for robots. Through this optimal energy flow and reduction in power losses, the efficiency in the robotics
EVs enhanced many folds.
Dijkstra’s graph search algorithm is used in Neaimeh et al. (2013), to find the minimum energy path to the destination with EVs.
An optimal range prediction for EVs is determined by finding the minimum energy route to the destination. Through this approach,
EVs can be optimally charged. This optimal charging minimizes the power losses and conserve the energy in the system. The impact
of EVs charging rates on smart grid power quality, harmonic losses, and transformer loading are studied in Moses et al. (2010). EVs
battery charger exhibit the high power nonlinear characteristics, therefore the power losses from current harmonics gradually de-
crease the power quality of smart grid. The decoupled harmonic load flow analysis is used in this scenario to test various charging
scenarios. The charging scenarios with minimum power losses are used to improve the power quality of the smart grid.

4.2. Optimization approaches used for minimizing the total cost of energy for users

There are various optimization approaches for EVs that focus on the objective of minimization of charging cost for the users. The
total cost of energy for users can be minimized by employing various optimization approaches. Not only the total cost of energy for
users is minimized but also the economic burden on the users can be minimized to great extent. The descriptions of various opti-
mization methods that are developed for the minimization of the cost of energy for the users are described below:
In Nguyen and Le (2014), a joint optimization algorithm has been proposed for the optimized EVs charging. Through this joint
optimization approach, a tradeoff is achieved through which not only the total cost of energy for users is minimized but also the user
comfort is also focused. Through this optimization algorithm, the energy demand during the peak load can also be minimized to
encounter the load variance. An optimized scheduling approach discussed in Wu et al. (2012b), schedule the load based on the
objective of minimization of energy cost for the users. The AU-based load scheduling, forecasts the energy prices for users and then
assigned the charging slots to the users following the day-ahead schedule. A dynamic dispatch algorithm is also developed to
minimize the total cost of energy for EVs owners.
A distributed EVs charging method employs the normalized normal constraint (NNC) method (Jiang and Fei, 2013), to solve the
optimization method for optimal charging of EVs. Through this method EVs are charged optimally in street parking lots. Lagrangian
decomposition is then used to make the optimization more and more scalable. In this case, large number of EVs can be charged in
street parking lots with minimum energy cost for EVs owners. EVs can also contribute in the power grid through V2G capability.
Another distributed optimal EVs charging approach than minimizes the total cost of energy for users use the global optimization
approach (He et al., 2012). Through this global optimization approach the power is optimally distributed to optimally charge EVs.
This global optimization approach, is so much flexible that it can handle the large number of EVs and also the random arrival of EVs.
By optimally managing the distribution system operator (DSO) (Hu et al., 2014), the energy cost for EVs charging can be minimized.
EVs are optimally charged by predicting the driving pattern. The thermal limits of transmission lines and the expected load on the
transformer is also minimized while reducing the total cost of energy for EVs charging. Through this optimal charging of EVs the
congestion between the DSO and the fleet operator is also minimized, which has resulted into the improved efficiency of the power
distribution system. Authors in Tan et al. (2014), have presented an optimal algorithm for EVs charging in smart grid. Through this
optimized EVs charging, the users can also contribute to the power grid and sell back the energy to the distribution power grid. The
energy is usually produced through the renewable distributed generator and then is stored in EVs. EVs then through the V2G
capability transferred the energy to the smart grid. Therefore, the load on the power grid can also be minimized through this
optimization approach. The demand curve is flatten after employing this optimization algorithm. The packetized EV charging is
discussed in Rezaei et al. (2014), in this approach EVs are optimally charged in the same manner as the bandwidth requirements in
the communications networks are satisfied. This packetized EVs charging is compared with the first-come, first-served scheduling
approach. This EVs charging approach seems to minimize the overall charging cost as now all EVs can equally access the limited
resources. The communication needs for this distributed EVs charging approach is also minimized to the great extent. Linear pro-
gramming is employed in minimizing the total cost of energy for EVs users in Tushar et al. (2014). With the help of smart micro-grids,
EVs charging problem is formulated as the mixed integer linear programming (MILP) problem and then its solution is obtained
optimally. The stability and flexibility of the micro-grids is also enhanced, because EVs can now transfer their excess energy using this
optimal charging approach. Through the integration of EVs in the smart micro-grids, the total cost of energy for other users can be
minimized to a great extent. Authors in Yang et al. (2014), use the risk-aware and day-ahead scheduling approach for charging of
EVs. In this optimal charging approach the charging problem is formulated as the two-stage stochastic linear programming, and then
this two stage linear programming problem is solved using the L-shaped method. This scheduling approach minimizes the total cost of
charging EVs using the importance sampling. With the increasing of EVs, the load on the power grid is also controlled using this linear
programming based EVs charging.
Convex optimization is employed in achieving the optimized EVs charging and minimizing the total cost of energy for users. The

398
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

charging locations or the aggregation agents are given the capability to charge EVs optimally. In Bessa et al. (2011), convex opti-
mization is used for deciding the optimal EVs charging cost by the aggregation agent. With the introduction of the smart grid, the
aggregation agent has gained much attention as the deciding agent for the prices of the electricity in the market. Convex optimization
is then used by the aggregation agents in forecasting the load on the smart grid. Not only the expected load in forecasted but also the
various ancillary services are also provided to enhance the performance of the smart grid. Authors in Wu et al. (2012b) also employ
the day-ahead scheduling approach in charging of EVs. In this approach, a dynamic dispatch algorithm is used by the aggregation
agent to minimize the charging cost. The power demand during the peak hours is also managed and controlled by the aggregation
agent. Both the centralized and decentralized EVs charging approaches are now used in controlling the charging cost and the gen-
eration cost (Gonzalez Vaya and Andersson, 2012). The convex optimization approach is employed in two stages to gain the lower
tariff at the peak hours. In the first phase, EVs batteries are treated as the virtual storage pools and then are dispatched using the
optimal power flow. In the second phase, the price of the electricity is broadcasted to the vehicles in the distributed manner in the day
a head schedule. Through this controlled EVs charging approach, the charging cost is minimized, and the resources are avoided to be
overloaded. In Gan et al. (2012), an optimized algorithm is proposed to optimally charge EVs and to minimize the cost of energy for
EVs users. The main focus is to shift the overall-load to the valley filling hours. This achieved usually during the night time. This
algorithm has the capability to converge within limited amount of time. This distributed EVs charging structure, does not require
considerable communications architecture. Hence, it can be implemented while residing within the scope of the limited resources. In
Wei et al. (2016), a normalized Nash equilibrium is employed to achieve the optimal EVs for minimization of charging cost for users.
Two stage charging method is applied is used while employing the convex optimization approach. While taking into consideration the
uncertainties of the renewable energy resources and congestion control, the charging problem is made convex. When the commission
fee is shared by the participants. It is observed that, when the charging and maintenance fees is shared by all the participants is a
convex function, then in this case a normalized Nash equilibrium is employed in minimizing the generation and charging cost of the
energy for users.
EVs can also be optimally charged using the model predictive control (MPC) approach (Di Giorgio et al., 2014). Total cost of
energy for users can be minimized while addressing the limitations of the smart grid, electricity market, technical constrains, and the
driving behaviors. Through this approach, EVs users not only witness in the minimization of the cost but also they can participate in
the demand side management (DSM). Authors in Dallinger and Wietschel (2012), presents the optimized EVs charging to minimize
the cost and load while taking into account the renewable energy resources. Various models are applied to optimally charged EVs in
smart grid. Usually, the stochastic model is used to determine the mobility behavior of EVs, optimization model to minimize the total
cost of energy for users, and market equilibrium model to predict the various electricity tariffs. However, the charging cost is defined
by estimating the price signals and mobility behaviors.
Optimized EVs based on the game theory optimization approach is discussed in the (Sheikhi et al., 2013). In this approach, the
smart meters, automatic chargers, and the charging schedule is optimized in such as coordinated fashion that the system seems to
conserve maximum energy. The generation capacity limitations and dynamic electricity prices at various time slots are optimized and
then are used to decide the start time and duration of EVs charging. The total cost of energy and the objective of the peak shaving is
achieved using this approach. The load on the semi-urban power grid and the cost of energy for EVs users can also be minimized by
employing optimization approach as discussed in Zhou and Cai (2014). This technique does not need any accurate predictions of the
demand, because it employs an online load and charging time and duration estimation. The game theoretic optimization approach is
distributed manner. This optimization approach is simple to implement and has resulted into the minimization of the charging cost
for EVs users. Game theory can also be implemented in optimizing the energy consumption profile of the building (Nguyen and Bin,
2012). Through the use of centralized EVs charging, the square euclidean distance between the instantaneous energy requirements
and the average energy demand of the building is taken into consideration and then is minimized. By employing the game theory, EVs
users can independently enhance the function utility.
Dynamic programming is also used in minimizing the total cost of energy for EVs users. Authors in Beaude et al. (2013), in-
troduced the centralized EVs charging method. Two methods namely, local and global optimization approaches are used to optimally
charge EVs to gain the desired objectives. In the local method, the potential impact of EVs charging on the residual distribution feeder
is estimated, while in the global optimization approach, EVs can a synchronously update their charging and discharging profile to
each other. In this way, the coordinated and well controlled EVs charging results into the lowering of overall cost of energy for users
of EVs. Both the global and local optimization problem is solved using the dynamic programming. Quadratic programming is used to
optimize EVs charging, while minimizing the charging cost and to gain the optimal power balancing (Sundstrom and Binding, 2010).
In this approach the optimization of EVs charging is divided into two phases. In the first phase, the linear approximation of is adopted
to determine the SOC of the batteries. In the second phase, the quadratic programming is used to gain the power balancing and to
minimize the total cost of energy for the users. In Bashash and Fathy (2014), the convex quadratic programming is used to optimize
EVs charging. The V2G capability is also addressed with the objective of minimizing the total cost of energy for EVs users. The
optimization method is used to estimate the circuit battery model. This circuit battery model is then used to develop the method that
overcome the energy losses.
Particle swarm optimization is used in Siddiqi et al. (2011a) for route optimization. Multi constrained optimal path (MCOP) is
used to define the optimal and short distance for EVs. Through the optimization of the route, EVs charging patterns and locations can
become deterministic. Route optimization through PSO, ultimately leads to conserve the energy that minimizes the total cost of
energy for users. Because, now EVs users can optimally schedule their EVs charging to the nearest charging locations.
Battery swapping stations (BSS) are also introduced to provide timely energy to EVs. To optimize the availability of the batteries
in the stock GA is employed (Wu et al., 2015). The alternative batteries are provided when needed and then the best strategy is used

399
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

to charge EVs batteries by developing the optimization scheme. Authors in Siddiqi et al. (2011b), used one of the optimization
method named as the simulated evolution to optimally find EVs charging locations. The optimal path is determined while considering
the optimal EVs charging objective as the main goal. In this way, the total cost of energy is minimized by reducing the charging
cycles.

4.3. Optimization approaches used for maximizing the aggregator profit

Various optimization approaches have been proposed to maximize the profits of the AU. The electricity market revolves around
this objective of enhancing the profits for the charging facility provider.
PSO is used to maximize the aggregator profits while optimally charging EVs (Hutson et al., 2008). EVs fleet in the parking lots are
optimally charged. The aggregator in these locations adopt the binary PSO to optimally charge EVs while aiming to gain the max-
imum profits. For this purpose, both the centralized and distributed approaches for EVs charging with larger parking lot area are
employed.
Authors in Sortomme and El-Sharkawi (2011), propose an algorithm to maximize the aggregator profits. This algorithm is de-
veloped in the context of uni-directional V2G capability. The uni-directional V2G capability requires very limited communication
structure. This aggregator profit maximization algorithm is implemented while considering the optional system load and the system
price in the centralized manner. The centralized nature of the algorithm, enable the aggregator to have a complete information
regarding the charging and discharging of EVs. The optimal EVs charging in multifamily dwellings (Qi et al., 2014), can be co-
ordinated for enhancing the profits for charging facility provider. Usually, the lagrangian relaxation method is used in the distributed
fashion to enhance the profits of the aggregator. The multistage mixed-integer problem is solved in the hierarchical way while
maintaining the system sustainability, flexibility, and protection of the customers protection.
Various design proposals for the aggregator are discussed in Han et al. (2010b) to achieve the maximum aggregator profit. The
optimal aggregator design is chosen while considering various limited constraints. Dynamic programming is then used for the optimal
charging of EVs. The optimization approach is introduced to maximize the profits for the aggregator, regulate the frequency, and to
support the V2G capability.
Convex optimization is used in Soares et al. (2014) to optimally charge EVs in real-time scenario. The aggregator profit is
increased by adopting the two methods in charging EVs. In the first method, the aggregator used the optimization methods for
charging of EVs and to maximize the profits. In the second method, the DSO adopt the optimal strategy to charge EVs and to improve
the performance of the smart grid. V2G capability, ancillary services, peak load shaving, minimization of cost for the users are
discussed in conjunction with the objective of maximizing the profit for the aggregator (Sortomme and El-Sharkawi, 2012a). In this
optimal charging of EVs, the optimization algorithm has been developed while taking into consideration the unusual EVs departure.
Unplanned EVs charging in context of enhancing the aggregator profit is also discussed in Sortomme and El-Sharkawi (2012b). The
proposed approach addresses the issues related to the bidirectional V2G capability. Combined biding proposals are introduced and
are then utilized by the aggregator to define the tariffs. In addition the optional load level and cost optimization problems are also
addressed with the bidirectional V2G capability.
Mixed integer linear programming (MILP) model is used in Jin et al. (2013) to optimally charge EVs for maximizing the ag-
gregator profits. The optimal charging problem with the objective of peak load shaving, reduction in the cost for charging, and
maximizing the profit for aggregator is solved using the LP. A communication protocol has also been developed to optimize the
communication between EVs and aggregators to define the tariffs. Authors in Vagropoulos and Bakirtzis (2013) employ the stochastic
optimization approach to gain the maximum profit from optimal EVs charging. The relation of the fluctuating energy demand and the
profit maximization is discussed in detail. Therefore, the authors have divided the fluctuating energy demand into the instructed and
uninstructed energy demands. Optimization approach is employed to solve these instructed and uninstructed energy demands with
respect to the profit maximization of the aggregator. To maximize the profit from the optimal charging of EVs, an optimization
approach has been developed that facilitate the aggregation agent (Bessa et al., 2012). With the help of optimization scenarios, the
aggregation agent can maximize the profit for the aggregation point, by predicting the future energy demand and optimally utilized
the secondary energy reserves. The bidding for the manual reserve is optimized in Bessa and Matos (2013) for increasing the profits of
the aggregator. For this purpose, two optimized algorithms have been proposed for EVs charging and controlling. Both optimization
schemes discuss the day-ahead and hour-ahead real-time scheduling for EVs charging.
Stochastic programming is used in Balram et al. (2013) to achieve the optimized EVs charging while addressing the issues of spot
price, expected load, voltage regulations. The proposed model deals with the problems faced by the electricity retailers. This sto-
chastic-based optimization model helps the retailers to add more profits by scheduling the large number of EVs for charging at an
optimal time and for an optimal duration.
Quadratic programming is also used to achieve the maximum profit for the aggregator from the optimal EVs charging (Deshpande
and Murali, 2013). The whole operation of the proposed optimization scheme is divided into two modules. In the first module, the
pricing for EVs charging is controlled in such a coordinated fashion, so that the aggregator achieves maximum profit. In the second
module, the scheduling is optimized to charge the large number of EVs in the parking lots.

4.4. Optimization approaches used for frequency regulation

Frequency regulation during EVs charging is an important objective addressed by various optimization schemes. Following are the
descriptions of various optimization schemes employed to regulate the frequency in the charging of EVs.

400
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Improved power production

and quality

Significant reduction in CO2

Enhanced Security

Smart grid

More robust power grid

Provision of optimized EVs

recharging

Support for renewable energy

resources

Fig. 4. Provision of Optimized EVs charging as the main function of the smart grid.

Dynamic programming is used for regulation of frequency (Han et al., 2010b) during the charging of EVs. Various design models
for aggregators have been proposed. The optimal design that helps to regulate the frequency is selected based on the cost function.
Two algorithms are proposed in Rotering and Ilic (2011) that are also based on the dynamic programming. The first algorithm
optimizes EVs charging by deciding the charging time and power flow. The second algorithm usually performs the frequency reg-
ulation to stabilize the optimal operation of the smart grid. Fig. 4 lists main contribution of smart grid. Various constraints such as
technical and impact of load on the smart grid are also taken into consideration during the charging of EVs.
Convex optimization as discussed in Sortomme and El-Sharkawi (2011), is presented to optimally charge EVs while regulating the
frequency of the aggregator. The unidirectional V2G capability is supported with the objective of frequency regulation and max-
imizing the aggregator profit. In this approach, the optimal charging based on the smart charging is introduced in which the convex
optimization enables to charge different rates depending on the various regulation facilities. Convex optimization problem is also
employed in Lin et al. (2014) to optimally charge EVs. This distributed EVs charging approach, used the gradient projection approach
in defining the scheduling method for EVs charging. The uncertainties that arise from the unexpected forecasting of the energy
demand and the revenues that arise from the providing the frequency regulatory services are also addressed through this convex
optimization-based approach.
Heuristic algorithm is used in Kang et al. (2013) to introduce the real-time charging of EVs while regulating the grid frequency.
The grid stability is achieved by incorporating the centralized EVs charging approach. This optimization approach also favors the
electrical energy storage in the batteries. Therefore, the charging and discharging cycles are stabilized while providing the frequency
and voltage regulation. Authors in McArthur et al. (2007), concludes the findings of the multi-agent system (MAS) workgroup in
relation to power engineering applications. The power engineering applications are modified in the context of the MAS for regulation
of the frequency of the aggregator and smart grid while optimally charging EVs. Different design directions have been proposed for
the architectures of the MAS. Small power systems that are powered by the petroleum generators, wind energy, and light energy are
controlled in the decentralized manner (Tokudome et al., 2009). In these power systems, the heuristic optimization algorithm is used
to regulate the frequency and voltage. Fluctuating frequency and voltages due to the various power sources are controlled in the
distribution manner. EVs when charged from the renewable energy source like the wind witness an immense fluctuation in the
frequency and voltage. Frequency variations, when EVs are charged from the wind energy source are regulated by the optimization
approach as discussed in Lopes et al. (2009). The V2G capability is also supported with the objective of contributing the energy in the
power gird. This wind supported, frequency regulating optimization approach operates in two phase. In the first phase, the opti-
mization is applied when EVs are integrated for charging, and in the second phase the frequency and the voltages are regulated. The
wind generator usually replaces the traditional power grid in an isolated locations such as the island. EVs with V2G capability are
employed to store the energy from the wind-power systems and then are utilized for various operations. Danish island follows the
same trend (Pillai and Bak-Jensen, 2010a) and are powered with the wind-power sources by using the V2G capability. Here, the
optimization approaches are utilized to stabilize the frequency irregularities. Authors in Liu et al. (2013a), discuss the charging with
frequency regulation (CFR) approach for optimal charging and frequency regulation in EVs. CFR is performed with wind-power
source integration through V2G capability. This approach is achieved with the help of the decentralized V2G capability (DVC). The
unidirectional V2G capability, with wind-power integration employs the CFR through the DVC, and achieves the desired optimal EVs
charging.
Stochastic optimization is introduced in Donadee and Ilić (2012), to regulate the frequency while optimizing EVs charging. The
deterministic optimization model is used in optimally estimating the revenues that an EV can earn by providing the ancillary services
such as frequency and voltage regulation. This optimization model also removes the uncertainties in the revenues that arise after
providing the frequency regulation. Frequency regulation by the V2G capability is now perceived as the most reliable compared to

401
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

the other conventional ways as has been shown in Pillai and Bak-Jensen (2010b). The wind power generation system with 48% and
65% contribution in the total power generation system in an island is analyzed for frequency regulation through V2G capability. The
V2G capability of EVs seems to show more reliable frequency regulation when compared to other traditional power generation
frequency regulation systems. Wind power fluctuations are also minimized with the decentralized EVs charging approach (Luo et al.,
2014a). This distributed EVs charging handle the wind power integration into the traditional power systems with frequency reg-
ulation capability. The V2G capability achieved from the optimal charging and discharging of EVs not only provides the frequency
regulation but also conserve the thermal units from the various harmful impacts of the wind power fluctuation.
Mixed integer linear programming (MILP) has been discussed in Jin et al. (2013a) to regulate the frequency of the aggregator and
smart grid. The concept of energy storage (ES) has been employed to optimally manage the load on the power system. A commu-
nication protocol has also been developed for communication between the smart grid, EVs, ES and the point of charging. This optimal
charging of EVs minimizes the load on the smart grid, regulates the fluctuating frequency, and increases the revenues for the
aggregator by employing effective charging scheduling of EVs and communication. Linear programming is also used in Mojdehi et al.
(2015), to optimally charge EVs and to obtain the various objectives such as frequency regulation. The energy cost function is usually
calculated to estimate the impact of battery degradation on the frequency regulation capability of EVs. The operating point and
optimal bidding problems are solved by employing the linear programming.
In Verzijlbergh et al. (2012), the optimization problem of EVs charging is solved by taking into consideration the consumers
behaviors, power generation capacity, and the expected load. These perspectives form the basis for modelling the charging of EVs in
such a way that, the frequency of the aggregator and distribution power grid remains stable in extreme condition. Author in Mullen
(2009), explores the various ways of the distributed EVs charging approach for regulating the frequency of the distribution power
systems. Various controllers for EVs charging have been proposed to equip EVs with the frequency regulation capability. A simulator
has also been proposed to study the various energy models for EVs and the impact of various energy levels on the batteries of EVs. The
simulation is then performed with the simulator to estimate the impact of optimal charging and the frequency regulation.
PSO is used to regulate the frequency of the distribution system (Celli et al., 2012). The aggregator in this case, is made capable to
determine the optimal charging and discharging of EVs. The aggregator in this case, takes the help of the V2G capability of EVs in
stabilizing the frequency of the power distribution system. The aggregator in this case is the main deciding factor for charging and
discharging of EVs that takes into consideration the mobility patterns, energy patterns and the bidding scenarios.
Quadratic programming as discussed in Han et al. (2011) is used to optimally charge EVs while regulating the frequency of the
aggregator. In this optimization approach, the energy constrains and the SOC is taken into the consideration while keeping the track
of the optimal power flow. The aggregator design is optimized in such a fashion that the frequency and voltage are regulated with
optimal EVs charging.
Game theoretic based optimization is used in Wu et al. (2012a) to optimally charged EVs. This distributed EVs charging approach
not only takes into consideration the optimal EVs charging but also consider the interaction between EVs and the their respective
aggregator. The optimization problem is solved through game theory, and frequency is seemed to be regulated in smart grid. This
approach also defines the smart price policy for EVs charging.

4.5. Optimization approaches used for voltage regulation

The inclusion of EVs in the existing power systems can result into the voltage fluctuation during peak hours. The voltage drops can
lead to severe hazards such as transformers dysfunctions. However, EVs can now contribute to the regulation of the voltage in the
distributed power systems. Various optimization approaches have been developed to regulate the voltage during EVs charging. The
detail of the some of the optimization approaches that take part in voltage regulation are discussed below:
Game theory based optimization approach is used in Beaude et al. (2013) to optimally charge EVs while taking into consideration
the voltage regulation. This distributed EVs charging approach, results into the voltage control on the distribution feeders and seems
to improve the efficiency of the distribution power systems. The Monte Carlo simulations have been performed to evaluate the
performance of the proposed system. The aggregator and EVs globally exchange their information asynchronously to update the
decision process. Game theory based multi-agent system has been proposed to optimally charge EVs in the distributed fashion
(Karfopoulos et al., 2011). In this proposed scheme, the multi-agent system not only coordinate the optimal EVs charging but also
control the charging and discharging of EVs batteries. The game theoretic based optimization approach is also employed in this case
while maintaining the interaction between the micro-grid and EVs for regulating the voltage.
Linear programming based optimal charging of EVs is discussed in Richardson et al. (2012b). Through this approach, the large
number of EVs can be charged by taking into consideration the charging capacity of the low voltage of distribution power grid. The
linear optimization to solve the charging problem and to minimize the cost for charging by regulating the voltage of the power
distribution system is employed in this method. A distributed linear programming based EVs charging approach is discussed in Ahn
et al. (2011). This is actually the two level optimization approach for EVs charging. In the first level, the voltage regulation is
achieved and in the second level, the frequency is regulated. Through this approach, a large number of EVs can be charged from the
low voltage distribution power network. The real-time management of EVs charging using the linear programming is also discussed in
the (Vagropoulos et al., 2016). Through this approach, the charging of EVs through aggregator is managed by defining the charging
priority, dynamic voltage regulation, and proper modeling of the battery.
Authors in Singh et al. (2012), used the fuzzy logic to optimize EVs charging. The controlled EVs charging is applied to regulate
the voltage and to minimize the load on the distribution system. The fuzzy logic controller (FLC), is used in two phases to achieve the
optimal charging of EVs. In the first phase, the charging station is controlled and in the second phase, the V2G capability is controlled

402
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Table 5
Different power levels used for EVs charging.
Type of power level Type of charger employed Voltage (VAC) Range Charging time Power flow Phase

Level 1 J1772 charge port 120 2 to 5 miles 1 h charging Unidirectional Power Flow Single
Level 2 J1772 charge port 208/240 10 to 20 miles 1 h charging Bidirectional Power Flow Single
Level 3 CHAdeMO and Tesla combo 240 50 to 70 miles 20 min charging Unidirectional Power Flow Three

employing the fuzzy logic.

Optimal power flow (OPF) is discussed in Abe and Saitoh (2014) to regulate the voltage during EVs charging. The distributed
generators (DGs) are optimally designed to estimate the loss during the charging and discharging cycles. The Ohmic losses are
reduced by using the lagrange multipliers for shifting the load. Table 5 shows the various power levels with regards to power flow,
range, and charging time. Voltage droop-based optimization approach is discussed in Huang et al. (2013) to optimally charge EVs and
to regulate EVs charging in one of the Danish island. Usually the voltage droop approach is used in combination with the reactive
energy management. The proposed approach is evaluated and is seem to show the improved performance in terms of voltage reg-
ulation and in reduction of cost. Another approach for optimal charging of EVs discussed in Mitsukuri et al. (2012) also takes into
consideration the reactive power control. The charger for EVs usually used the self-commuted inverter for charging EVs in an optimal
way. The main contribution for using the this kind of charger is to control the reactive and active power for optimal charging of EVs.
In this case, the Monte carlo simulations is performed to investigate the optimal charging of EVs.
Heuristic method for optimal charging of EVs is discussed in Cao et al. (2012). In this method, the peak and valley time periods are
partitioned to optimally set the time of use (TOU) price for the charging of EVs. The optimization model based on the heuristic model
is then developed that not only address the voltage regulation objective of EVs, but also minimizes the charging cost for EVs charging.
Genetic algorithm-based EVs charging is discussed in Jiménez and García (2012) to regulate the voltage of the distributed sys-
tems. The three phases of optimal power flow are employed and then the optimization problem is solved using the genetic algorithm.
A voltage source converter for EVs fleet is developed to optimally charge EVs fleet. The voltage is usually regulated with the help of
the this voltage source converter. The voltage is also regulated during the charging of electric bus using the genetic algorithms
(Herrera et al., 2015). In this case, the electric buses with the combination of supercapacitors and battery storage are used. Usually, a
power splitter is employed to distribute the power between the supercapacitor and the battery. Through this approach, the opti-
mization problem is solved with the help of genetic algorithms, and the results are presented in pareto front. The voltage is not only
regulated but the electric vehicles have witnessed in the minimization of the cost for EVs charging.
Authors in Cheng et al. (2015) propose the two phase method for regulating the voltage of the distribution system. In this optimal
charging of EVs, online load tap charging of EVs has been introduced to regulate the voltage and to provide the V2G capability.
Table 6 compares the benefits of optimized and un-optimized EVs charging. In this two phases of EVs charging, the online tap
charging is controlled in the first phase, and in the second phase the fluctuations by the transformers are controlled. Stochastic
modeling approach is introduced in Fazeli et al. (2015), to optimally charge EVs and to regulate the voltages in the low voltage
distribution network. In this bottom up EVs charging approach, the inclusion of various EVs is estimated in low voltage distribution
system. The stochastic modeling is then employed in the bottom up way and the voltage is regulated during the optimal charging of
EVs. In Bai and Qiao (2015), the stochastic behavior of the aggregator is modelled as the robust optimization problem. Then this
optimization problem is solved using the mixed integer quadratic programming. In this optimal charging of EVs, the aggregator is
usually used as the main agent in regulating the voltages of the distribution power grid. The impact of EVs charging on the real
distribution power grid is evaluated in Saberbari and Saboori (2014). The impact on power grid with respect to power losses and
voltage drop is estimated with the help of quadratic programming. Then the charging of EVs is scheduled in such a fashion that the
voltage is regulated, that helps in minimizing the negative impacts on real distribution grid.

4.6. Optimization approaches used for supporting the renewable energy sources for charging of EVs

The depleting energy resources compel the researchers to come out with the various solutions to charge EVs in an optimal way.
Various renewable energy resources have been employed for charging of the increasing number of EVs. Various optimization ap-
proaches have been used to support the inclusion of the renewable energy resources for optimal charging of EVs. The description of
the various optimization approaches for EVs charging using the direct or indirect renewable energy resources is given as follows.
Quadratic programming is used in Zhao et al. (2015) to optimally regulate the inclusion of the wind power in the distribution
power grid for charging of EVs. The impact on the distribution power grid due to the increase in the number of EVs and fluctuation in
the wind power is assessed and then this optimization problem is solved with the help of the quadratic problem. In this case, the
battery swapping stations are provided with the wind power and then EVs are optimally charged using the optimization approach.
Quadratic programming is also employed to optimize the solar polar in the local distribution power grid (Ratnam et al., 2013). EVs
battery storage and the solar power are then optimize in a way that EVs are charged at the lower rate. The energy time shifting for the
optimal charging of EVs and the impact of the battery storage on the distribution power grid is assessed using the quadratic pro-
gramming.
Authors in Lopes et al. (2009), assess the high intermittent wind power impact on an isolated power grid. This (wind) renewable
energy source is considered to be the main factor in fluctuating the voltage and frequency of the distribution grid. The impact of the

403
 M. Amjad et al.

Table 6
Comparative view of optimized and un-optimized EVs charging.
EVs charging Main advantages Drawbacks Power level used EVs charging Approach Communication needs V2G

Un-optimized EVs EVs can be charged at any time and at any place, Power losses, voltage and frequency drop, increase in Level 1 is used for EVs are charged whenever No No
charging there is no need of any sophisticated infrastructure overall load, degrade the distribution network charging of EVs they are parked
for charging, EVs can be charged at minimal rate efficiency, decreases the aggregator profit, and do not
support the renewable energy integration

404
Optimized EVs Reduce power losses, minimize total cost of energy Need well-sophisticated and costly charging Level 2 and 3 are Centralized and A well-designed Yes
charging for users, maximize aggregator profits, voltage infrastructure and well-designed communication used for charging distributed EVs charging communication
regulation, frequency regulation, support for infrastructure of EVs approaches are used infrastructure is needed
renewable energy sources for charging, avoid
distribution network issues, and minimize the peak
load of EVs
Transportation Research Part D 62 (2018) 386–417
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

variation of the wind power is assessed from two perspectives. First, when EVs are in charging mode and secondly, when EVs are
contributing in regulating the frequency of the distributed power grid. The whole optimization problem is then solved using the
heuristic approach. The wind power fluctuations can also be minimized with the help of graph search algorithm (Luo et al., 2014a). In
this distributed EVs charging approach, the renewable energy resources are optimized in the system by EVs. EVs minimize the wind
power fluctuations with the help of the communal virtual price signals, by solving the optimization problem with the help of the
graph search algorithms.
An optimized algorithm has been proposed in Fan (2012), to charge EVs using the renewable energy resources. Usually, the
congestion in the network traffic of the smart grid is taken into consideration. The proposed optimization approach in this case
enables EVs to charge according to various charging rates depending on EVs preference. This distributed EVs charging approach also
estimates the impact of the renewable energy source contribution in minimizing the load on the smart grid. The event-based opti-
mization is employed in Huang et al. (2014a) to solve the optimization problem of fluctuating available renewable energy source.
This optimization problem is formulated as the Markov decision process and then the event-based optimization is used to chart out
the optimal charging approach for EVs. The renewable energy source in this case is wind which is the highly fluctuating power source
contributing in the distribution network. The renewable energy source is highly unpredictable in its availability, therefore to utilize
the renewable energy for EVs charging, lyapunov optimization is used (Jin et al., 2014; Jin et al., 2013b). When EVs are charged from
the renewable energy source, this results into the minimization of the charging cost for EVs users. This whole optimization problem is
also formulated as the stochastic and then solved using the lyapunov optimization. An optimal scheduling approach for the charging
of EVs at the charging stations is discussed in Zhang et al. (2014a). The charging stations are provided the energy from the renewable
energy source and the power grid. An optimization approach is used to minimize the waiting time for the charging of EVs. The
optimization problem encompasses the availability of the renewable energy source, the charging demand queue length, the arrival of
EVs fleet for the charging at the charging station, and capacity of the storage batteries. This optimization problem is formulated using
the stochastic optimization and then solved using the Markov decision process. Distributed energy resources (DER) for controlled EVs
charging are coordinated using various optimization approaches (Jin et al., 2012). The stochastic approach is used to investigate the
optimal integration of the renewable energy and to manage the DER. The voltage regulation, charging cost minimization, and peak
load shaving is witnessed when the DER are optimally controlled and scheduled using the stochastic approach.
The solar photovoltaics (PV) systems can also be integrated with the charging stations for optimal charging of EVs (Lu et al.,
2014). In this situation, the renewable energy sources are utilized immediately without using the storage batteries. The multi-
objective optimization approach used in this case takes into consideration the electricity prices, and the availability of the renewable
energy source. EVs can also be used to transfer the renewable energy to the location where the renewable energy source is not
available (Arikan et al., 2015). In this case, the optimization problem is solved using the linear programming. The optimal charging of
the electric buses from the renewable energy source-based charging station is optimally scheduled and then the storage of these buses
are uses to transfer that energy. This ultimately results into the reduction of the electricity prices at the location where the renewable
energy sources are scarce.
Genetic algorithm-based optimization is used to optimally charge EVs from the renewable energy source (Liu et al., 2014). In this
coordinated EVs charging approach, the peak valley load is minimized using the probabilistic model. Probabilistic model uses the
output from the wind and PV source and then estimate the differences in the peak valley. The whole optimization model is then
solved using the genetic algorithm. The genetic algorithms are also employed in deciding the optimal size and location of the wind
power based charging stations for the electric buses (Yan et al., 2013). During the construction of the renewable energy based
charging stations, the urban environmental conditions are also taken into consideration. The impact of the optimal designed charging
station on the distribution network is then assessed using the genetic algorithms.
The V2G and G2V capability of EVs in conjunction with the renewable energy source is also optimized using the optimization
approach (Nguyen et al., 2015). The certainty equivalent adaptive control (CEAC) approach is used to control the charging and
discharging of EVs. The renewable energy source, G2V, and V2G capability is handled by employing the optimization approach. The
impact of the integration of the renewable energy source into the existing power grid and the charging of EVs have been extensively
assessed (Richardson, 2013). The economic, environmental, and social impacts of the integration of the renewable energy sources
into smart grid for EVs charging is assessed from many aspects. It is concluded that the wind power is integrated more compared to
the PV or solar energy in the system. Various approaches for minimizing the fluctuations in the frequency and voltage due to the
unpredictable behavior of the renewable energy source are also presented with much detail. The V2G capability of EVs is usually used
to stabilize the power grid. Various optimization approaches for tapping the energy in EVs are also provided in Kempton and Tomić
(2005a). Various steps for implementing the V2G capability while using the renewable energy source during the charging of EVs are
discussed. The V2G capability comes out to be more economical in transportation of the electricity to the places where the renewable
energy source is not available. The renewable energy-based fast charging stations have been optimally controlled using the various
optimization approaches (Mauri and Valsecchi, 2012). EVs are usually charged from the wall boxes of these fast charging stations.
Various optimization approaches are usually employed in coordinated the charging of large number of EVs. The fluctuations in the
energy supply of fast charging stations are handled using the optimization approaches, especially the peak load is shaved in the smart
grid. Authors in Costa et al. (2014), discuss the various benefits of integrating EVs charging and renewable energy sources. Various
limitations that arise after integrating the renewable energy source into the existing power grid are addressed by using the robust
optimization approaches. The renewable energy sources are usually quantified in case of wind and PV sources to optimally charge
EVs in the distributing grid. A multi-time scale optimization has been discussed in Huang et al. (2015) to integrate the wind power
into the micro-grid for EVs charging. A predictive control approach based on the Markov decision process has been introduced to
optimally charge EVs from the fluctuating wind power generators.

405
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

4.7. Optimization approaches used for avoiding the distribution network Issues

Inclusion of EVs poses various threats to the distribution network. The various issues such as harmonics, power losses, breakdown
due to the thermal limits violations, excessive load during peak hours, and voltage fluctuations can result from the uncontrolled EVs
charging. Various optimization approaches have been introduced in the literature to mitigate the issues related to the distribution
network. The description of the some of the optimization approaches used during EVs charging that minimizes potential threats to the
distribution network are discussed below:
Artificial immune system has been used in Oliveira et al. (2013) for minimization of load and regulating the voltage in EVs. An
optimal charging period for EVs is decided and then the load is distributed along that period. In this optimal charging of EVs,
capacitors are also used to overcome the electric losses that occur during the charging of EVs. Maximum sensitivities selection (MSS)
optimization is used during the charging of EVs (Deilami et al., 2011) to optimally control the distribution power grid. The per-
formance degradation, expected load, and various harmonics that occur after inclusion of EVs into the smart grid can be controlled
using the real-time smart load management (RT-SLM). Then this optimization problem is solved using the MSS optimization ap-
proach. Through this approach, not only the load on the smart grid is minimize but also factors that contribute to the degradation of
the performance of the smart grid are also minimized. Heuristic algorithm has been employed in Singh et al. (2010) to minimize the
impact of penetration and charging of EVs into the smart grid. During this approach, the quality of the power flow, frequency and
voltage regulation, minimization in the generation cost, and the reduction in the power losses are achieved. The actual data from the
charging and discharging locations of EVs are simulated and the potential impact on the power grid and the transmission lines are
optimized to regulate the voltage and frequency of the distribution grid. Another heuristic approach to optimize EVs charging load on
the smart grid is proposed in Ruiqin and Zhongjing (2015). Through this approach, the various characteristics of EVs charging is used
to evaluate the whole charging process. Usually, the start time of EVs charging, EVs battery, and initial SOC of EVs are optimally
controlled to improve the daily load profile of the distribution network.
Game theoretic-based optimization to minimize the impact on the power grid is proposed in Zhou and Cai (2014). Through this
optimization approach, EVs are charged in the distributed manner. Through this optimization approach, the arrival and charging
time of EVs is predicted online. The bus congestion and the frequency fluctuations that impede the normal operation of the dis-
tribution grid is minimized through this distributed game theoretic-based optimal charging approach. Game theoretic optimization is
also used in sharing the charge between EVs (Dutta and Boulanger, 2014). Through this approach, the potential load on the dis-
tribution grid is minimized during the peak hours. The Nash bargaining approach is used to calculate the incentives achieved by EVs
participating in the charge sharing network. Usually, in this approach, EVs can share the charge through the inductive power transfer.
Convex optimization approach is used during EVs charging while residing within the various technical limitations of the dis-
tribution grid (Soares et al., 2014). EVs are charged through two methods while employing this optimization approach. In the first
method, the control at EVs aggregator is used, while in the second method, the DSO is controlled through the centralized approach. A
local control approach for EVs charging using the convex optimization is also discussed in Richardson et al. (2012a). The local
charging approach uses the test distribution network and takes into consideration various distribution network issues. This local
charging of EVs is then compared with other centralized EVs charging and proved to be the most efficient and optimal for charging
EVs using the convex optimization.
Authors in Lopes et al. (2011), provides the conceptual model for the integration of EVs into the distribution network. With
reference to the various optimization approaches, the impact of the integration of EVs fleet on the distribution network is analyzed
from various perspectives. The optimal operation of the distribution network and the electricity prices with reference to the elec-
tricity market is also assessed using various heuristic algorithms.
Dynamic optimization has been used in Iversen et al. (2014) to optimally charge EVs. The distribution network issues such as
voltage drop, power losses, and harmonics are addressed in this distributed EVs charging approach. To conserve the maximum
energy, various driving patterns have been developed and assessed. The Markov decision process is then used to evaluate various
driving patterns. Dynamic programming is also used in charging EVs where the traffic congestion is the contributed factor
(Pourazarm et al., 2014). The homogeneous charging nodes are then grouped into the inhomogeneous subflows. These subflows are
then charged using the dynamic programming. This optimal charging of EVs resolves the issues of the overloading on the distribution
network during the peak hours by predicting the charging time and duration. The optimal charging framework for battery swapping
stations (BSS) has also been developed using the dynamic programming (Sun et al., 2014). In this approach, the network queuing
model is developed to control the charging operation of batteries in BSS. The optimal charging of the batteries in the BSS minimizes
the potential load and generation cost of the distribution network. The optimal power flow in the charging operation of the BSS with
QoS is also achieved using the dynamic programming.
Heuristic algorithms are also used to solve the optimization problem of the controlled EVs charging. Authors in Lopes et al.
(2007), analyzed the integration of the distributed generators into the existing electric systems. The impact on the transmission
system, protection of the distribution system, and the provision of the ancillary services by the integration of the distributed gen-
erators is assessed using the heuristic approach.
Quadratic programming is also used by EVs charging units while considering the various distribution network constraints
(Sundstrom and Binding, 2012). The charging units usually consider the voltage and power constraints of the existing power units
and then solve the optimization problem of the controlled charging of EVs using the quadratic programming. Congestion in the power
network is also minimized with the help of the quadratic programming (Huang et al., 2014b). Distribution locational marginal pricing
(DLMP) is used to overcome the congestion, and to stabilize the power flow in the distribution grid. In this centralized EVs charging
approach, the karush-kuhn-tucker (KKT) conditions are applied in this optimization approach to optimally charge EVs and to solve

406
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

the various issues related to distribution grid. Various issues related to the micro-grid using the quadratic programming are also
addressed in Cai et al. (2013). During this charging and discharging of EVs, day ahead scheduling is proposed to minimize the
distribution network losses and to optimize the economic operation of the micro-grid. Through discharging, EVs can contribute in
maintaining the voltage fluctuations and provide the ancillary services. This optimization problem is then solved using the quadratic
problem. EVs are also optimally charged from the home energy control box using the quadratic programming (Mets et al., 2010). This
optimized EVs charging approach results into the minimization of the peak load and various voltage fluctuations. This quadratic
programming-based optimization approach is compared with various business as usual scenarios. Serial quadratic programming for
the controlled EVs charging is discussed in Zhang et al. (2012). The whole power system is optimized such that EVs are charged in an
optimized way. In this approach, a statistical model is presented to predict EVs charging load on the existing power system. Four
different scenarios of EVs penetration into existing power system is modeled and the optimization problem is solved using the
quadratic problem.
The impact of the integration of EVs into the existing power system is also discussed in Hu et al. (2012). Various management and
planning issues such as increasing the generation capacity, optimizing the grid operation, controlling and planning the charging
facilities have been discussed with respect to various charging scenarios. The various optimization approaches have also been dis-
cussed to control the overall charging process. Authors in Qian et al. (2011), simulate the four different EVs charging approaches to
predict the future changes in distribution systems and in the electricity tariffs. A stochastic model is used to predict the expected EVs
penetration into the distribution system. The four charging scenarios are the unregulated home charging, un-optimized off-peak
domestic charging, controlled charging of EVs from the residential power grid, and uncontrolled public charging. In addition to four
charging approaches, in Freire et al. (2010) three charging approaches have been discussed for optimally charging EVs using various
optimization methods. These three charging approaches are dumb charging, smart charging, and V2G capability. The integration of
the renewable energy source with these three EVs charging approaches have also been analyzed using the heuristic algorithms.
Table 7 shows the optimization objectives with their main optimization approaches and limitations.
Particle swarm optimization is discussed in Tian et al. (2012) for charging EVs battery. In this multi-objective dispatching
strategy, the impact on the load curve due to the penetration of EVs into the distribution grid is estimated. Then the results of the
single-objective optimization and multi-objective optimization are compared with one another. The multi-objective optimization-
based on the particle swarm optimization seems to show better performance in conserving the energy while optimally charging EVs.
A genetic algorithm-based optimization approach is used in Shaoyun and Liu (2012) to obtain the optimal time period for load
shifting for EVs charging. Monte carlo simulation is performed to gain the load curve for the charging periods. Usually, this approach
is used to fill the load valley of the distribution network that results from the integration of EVs charging in the system. Optimized
location scheme for electric charging stations (OLoCs) (Mehar and Senouci, 2013) uses the genetic algorithm in finding the location of
the charging stations for EVs. Using this approach, the optimal number and location of the charging stations for EVs can be found. The
whole optimization problem is solved using the heuristic approach that employs the genetic algorithm. The charge patterns of EVs
can also be determined using the genetic algorithms (Bashash et al., 2011). Using this optimization approach, the pareto front
charging pattern of EVs is obtained and the total cost for the electricity and the fuel cost for generating the electricity is estimated.
Linear programming is used in Rahimi et al. (2014) to estimate the possible capacity limitations of distribution network due to the
integration of the large number of EVs into the existing system. Volt/var optimization-based on the linear programming is used to
estimate the impact on the distribution network. Various operating conditions of the distribution network is taken into consideration
for charging of EVs. An optimal charging scenario is then chalked out while using the linear programming.
Fuzzy logic can be used in optimal charging of the driverless EVs from the auto charging stations (Luo et al., 2014b). The
driverless EVs navigate the nearest charging station through vector field histogram (VFH). This local navigation tool directs the
driverless EVs to charge optimally from the auto charging station using the fuzzy logic.

4.8. Optimization approaches used for minimizing the peak load

The integration of the large number of EVs into the existing distribution network increases the load on the distribution network.
The peak load can result into various harmonics, voltage and frequency fluctuations, overloading the transformers, and increases the
power losses. Various optimization methods have been introduced to maintain the load into the optimal level. Below is the de-
scription of the different optimization schemes used to optimally charge EVs while taken into consideration the objective of mini-
mizing the peak load.
Authors in Gan et al. (2011), use the decentralized EVs charging approach to optimally charge EVs. In this approach, EVs charging
problem is formulated as the optimal control problem. This optimal control problem is then solved using the distributed optimization
algorithm in an iterative way. This results into the minimization of the peak load and increases the operational capacity of the
distribution network. The optimal charging of EVs using the distributed approach is also discussed in Ma et al. (2010). Nash certainty
equivalence principle is used to schedule the charging of large number of EVs. In this heuristic optimization scheme, the compu-
tational complexity of the proposed scheme is also taken into consideration and the optimization objective of minimizing the peak
load is achieved. Another heuristic optimization approach using the distributed EVs charging approach is discussed in Torabi and
Gomes (2014). Through this approach, a large number of EVs in the form of group can be charged at an optimal rate. This scheme also
takes into consideration the electricity tariff schemes, various charging patterns, users preferences. This decentralized EVs charging
approach ultimately flatten the load profile.
EVs can also be charged using the centralized EVs charging approach. In Le et al. (2015), the stochastic optimization-based
centralized EVs charging approach is proposed. This approach minimizes the peak load and charging and discharging cost for EVs.

407
 Table 7
Optimization objectives with their optimization and charging approaches.
Optimization objectives Main optimization approaches involved Constraints Charging Approach Main process involved Main solver tools Constraint
M. Amjad et al.

Violation

Reduce power losses Quadratic Programming (Clement-Nyns et al., Power resource Both centralized and TCOPF, SQP, RTSLM, Matlab, JADE (Java), Under voltage
2010; Bashash and Fathy, 2013; Zhu et al., 2004; minimization and power distributed MPDIPA, and PSO- and PowerACE
Brembeck and Ritzer, 2012) cost CFA
Convex Optimization (Sortomme et al., 2011;
Pourabdollah et al., 2013; Murgovski et al., 2013)
Particle Swarm Optimization (Grau Unda et al.,
2014; Dixit and Roy, 2015; Ko and Jang, 2013)
Dynamic Programming (Clement-Nyns et al., 2010;
Li et al., 2012)

Minimize total cost of energy Convex Optimization (Bessa et al., 2011; Wu et al., Charging availability and Both centralized and NNC, MILP, MCOP, Matlab, Simulink and Phase unbalances
for users 2012b; Gonzalez Vaya and Andersson, 2012; Gan load distributed and MPC PowerACE
et al., 2012; Wei et al., 2016)
Quadratic Programming (Sundstrom and Binding,
2010; Bashash and Fathy, 2014)
Linear Programming (Tushar et al., 2014; Yang
et al., 2014)
Game theoretic (Sheikhi et al., 2013; Zhou and Cai,
2014; Nguyen and Bin, 2012)

Maximize aggregator profits Heuristic (Sortomme and El-Sharkawi, 2012a; Electricity balance and In most cases the distributed MILP, and binary PSO Matlab, Matpower, PSS/ Frequency
Sortomme and El-Sharkawi, 2012b; Jin et al., 2013; rate EVs charging approach is E, and GAMS fluctuations

408
Vagropoulos and Bakirtzis, 2013) used
Dynamic programming (Han et al., 2010a; Han
et al., 2010b)
Convex optimization (Sortomme and El-Sharkawi,
2011; Soares et al., 2014)

Frequency regulation Dynamic Programming (Han et al., 2010b; Rotering Minimization of the power Both centralized and MAS, CFR, DVC, and DigSilent, PSS/E, and Power losses
and Ilic, 2011) resources distributed MILP PowerACE
Convex optimization (Sortomme and El-Sharkawi,
2011; Lin et al., 2014)
Graph search (Luo et al., 2014a)
Particle swarm optimization (Celli et al., 2012)

Voltage regulation Fuzzy logic (Singh et al., 2012) Minimization of the power Both centralized and FLC, and OPF Matlab, Simulink Power losses
resources distributed
Linear programming (Richardson et al., 2012b; Ahn
et al., 2011; Vagropoulos et al., 2016)
Genetic algorithms (Jiménez and García, 2012;
Herrera et al., 2015)
Quadratic programming (Bai and Qiao, 2015;
Saberbari and Saboori, 2014)

Support for renewable energy Stochastic optimization (Huang et al., 2014a; Jin Batteries limitations and Centralized but in some DER, CEAC, MDP, and Matlab, GAMS, and Frequency
sources for charging of EVs et al., 2014; Jin et al., 2013b; Zhang et al., 2014a) charging capacitors cases the distributed MILP Matpower irregulation
Quadratic programming (Zhao et al., 2015; Ratnam
et al., 2013)
(continued on next page)
Transportation Research Part D 62 (2018) 386–417
 Table 7 (continued)

Optimization objectives Main optimization approaches involved Constraints Charging Approach Main process involved Main solver tools Constraint
Violation
M. Amjad et al.

Genetic algorithms (Liu et al., 2014; Yan et al.,

2013)

Avoid distribution network Quadratic programming (Sundstrom and Binding, Peak load and power rates In most cases the distributed MSS, RT-SLM, KKT, Matlab phase-unbalance
issues 2012; Mets et al., 2010; Zhang et al., 2012; Huang EVs charging approach is OLoCs, and VFH
et al., 2014b; Cai et al., 2013) used
Genetic algorithms (Shaoyun and Liu, 2012;
Bashash et al., 2011; Mehar and Senouci, 2013)
Dynamic programming

Minimize the peak load Quadratic programming (Mets et al., 2010, 2012) Minimization of the power Both centralized and mSM and MILP Digsilent PowerFactory under-voltage
resources distributed and Matlab
Dynamic programming (Feng et al., 2013)
Genetic algorithms (Mehboob et al., 2014)
Convex optimization (Sojoudi and Low, 2011)

409
Transportation Research Part D 62 (2018) 386–417
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

For this scheme, the accurate predictions about the energy demand, renewable energy source contribution in the existing system, and
EVs charging patterns are not strictly made.
Quadratic programming is used to minimize the peak load and to flatten the overall load profile (Mets et al., 2010). In this optimal
charging of EVs, home energy control box is used to implement the two charging strategies. The two charging strategies are the local
charging strategy and the global charging strategies. Authors in Mets et al. (2012), proposed two optimal charging approaches to
minimize the peak load and to conserve the maximum energy in the system. The first approach is based on the quadratic pro-
gramming and the second approach is the controlled charging of EVs based on the market prices. Then these two approaches are
compared with the uncontrolled EVs charging. It shows that the controlled optimized charging of EVs minimize the peak load and
power losses compared to uncontrolled charging of EVs.
A dynamic optimized approach for EVs charging based on the particle swarm optimization is proposed in Yuqing et al. (2014).
Through this optimized EVs charging approach, usually the peak valley gap is minimized. Both the controlled and uncontrolled EVs
charging approaches are implemented in micro-grids. Optimized EVs charging approach based on the PSO seems to minimize the
peak load compared to the uncontrolled EVs charging approach.
A distributed EVs charging approach is discussed in Mierau et al. (2012) to optimally charge EVs. An optimization algorithm
based on the heuristic approach is proposed and implemented in the small device in EVs. Mobile smart meter (mSM) is usually used in
EVs that support the proposed optimization algorithm. Through help of this mSM, the optimal charging duration is assessed to
minimize the load on the power grid. The overall communications needed for EVs charging is controlled through this installed mSM.

5. Open issues, challenges, future research directions, and limitations of this study

The optimized charging of EVs while adhering to the limited resources also pose various challenges. In this section, we encompass
the various future research directions, challenges, and open issues that are involved in the charging of EVs.

5.1. Standardization of EVs charging communication and control structure

To control the optimized EVs charging of EVs, a sophisticated communications infrastructure is needed. Usually, a wireless
communication infrastructure is employed to control and monitor EVs charging. To achieve the optimum performance and to avoid
any bottlenecks, a well-defined standardization of EVs communications technologies as well as other EVs charging infrastructure is
required. Authors in Ahmad et al. (2017) have provided a comprehensive review of different charging technologies related to EVs
charging. In this study, different standards related to the EV charging has been considered. The standards relate to the wireles power
transfer to charge the EVs have also been considered. Integrating the state-of-the art technologies of communication such as full-
duplex, energy-harvesting, and cognitive radio networks with EVs communication architecture should also be explored in future
(Amjad et al., 2018).
The communication infrastructure involved for the EVs charging has still to be explored with the other state of the art tech-
nologies such as energy-harvesting, green communication, wireless sensor networks, and cognitive radio networks (Amjad et al.,
2017b; Amjad et al., 2016; Amjad et al., 2017a; Umer et al., 2016a). The integration of these concepts have already been employed in
the smart cities. Therefore, the integration of optimized EVs charging infrastructure in smart cities should be tested with the above
mentioned state-of-art concepts.

5.2. Wireless power transfer

To achieve the efficiency in the transportation technology, the charging of EVs through wireless power transfer should also be
considered. With the help of wireless power transfer, EVs can be charged without interrupting their driving schedule. The in-
troduction of the wireless power transfer also minimizes the need of smart vehicle parking lots where EVs are usually parked for
charging. Wireless power transfer for EVs charging while considering different state of the art concepts such as green communications
have been studied (Lu et al., 2018).

5.3. EVs charging and scheduling in smart cities

Smart cities have gained much attention due to the provision of smart solution for the existing problem of urbanization. Smart
cities have also been designed to provide the smart charging solution for EVs. However, the optimized EVs charging and scheduling of
large number of EVs in the smart cities have not been explored in detail. This study (Brenna et al., 2017) has discussed various aspects
of EVs charging in smart cities. The distributed EVs charging infrastructure has been especially discussed with different dimension of
involved communication infrastructure.

5.4. Vehicle-to-anything

V2G have been explored in detail while studying the optimized EVs charging. The concept of vehicle-to-everything, also helps in
transferring the electricity from the vehicle to everything in our environ. With the help of vehicle-to-everything approach, different
energy-scarce equipment can be charged or their energy needs can be fulfilled from EVs. This idea needs an extensive evaluation
while considering the different optimization approaches.

410
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

5.5. Security and privacy issues

EVs can be charged by using either the centralized or distributed EVs charging approach. In the centralized EVs charging ap-
proach, the data of EVs driving pattern, state of the charge, and the energy needs are demanded and collected by the central entity.
This pose a serious security threat. Any security breach at the central entity of EVs charging can leak the data of several EVs customer.
This demands an efficient security and privacy mechanism at EVs charging sites. Current research work on security and privacy for
EVs charging does not include in-depth analysis of different security threats. Security and privacy related issues for EVs charging has
been highlighted in Li et al. (2017). In this study, Portunes+, a privacy preserving, authentication scheme has been studied for the
EVs charging. However, this scheme has not been tested to various centralized, distributed, and hybrid EVs charging approaches.

5.6. Smart grid and EVs charging

The optimization approaches used for EVs charging can minimize the power losses and peak load on the smart grid. However, the
unchecked inclusion of EVs into the existing power grid or futuristic smart grid can pose serious threat to the efficiency of the smart
grid. Therefore, the integration of the various renewable and non-renewable energy sources into the power grid should also be
optimized with the increase number of EVs.

5.7. Limitations of this study

This study has encompassed the various dimensions related to the EVs charging. However, below is the description of the various
limitations of the this study. These limitations actually pave the way for future research directions related to the EVs charging.

5.7.1. Forecasting the EVs charging patterns

Predicting the charging pattern or behavior is considered to be an important factor to enhance the efficiency of a charging
infrastructure. Forecasting the EVs charging patterns also helps to select the optimal charging time, that will ultimately reduces the
cost of charging. Charging at the peak hours usually can burden the existing power sources, therefore with the forecasting of EVs
charging patterns, optimal charging time and duration can be selected. Authors in Hadi Amini et al. (2016) have considered the
forecasting of EVs charging behavior in distributed charging infrastructure. With the help of this forecasting strategy, the cost and
time involved for EVs charging has been reduced. Due to broad explanation of optimization techniques for centralized and distributed
EVs charging, forecasting of EVs charging has not been surveyed in this survey. However, as a future work, we will tend to explore
this topic in more detail.

5.7.2. Hybrid EVs Charging

This survey presents an in-depth review of different optimization approaches used for centralized and distributed EVs charging.
Hybrid approach of EVs charging is also considered an optimal way of charging the EVs. The limitations of centralized and distributed
EVs charging can be addressed with the help of hybrid EVs charging. However, due to the availability of very limited work on hybrid
EVs charging, we have not discussed this in much detail. However, this topic can be of great interest for the researchers in future.

5.7.3. Optimal capacity sizing

With the help of optimal capacity sizing of charging batteries for EVs, the efficiency of charging batteries can be increased.
Authors in Ugirumurera and Haas (2017) have explained the main contributions of employing the optimal capacity sizing approach
for EVs batteries. We have tried to encompass the optimal capacity sizing with the review of different optimization schemes, however
still this topic needs a thorough understading while considering the state-of-art work on EVs charging.

5.7.4. Heterogeneous EVs charging

EVs can be charged by employing the centralized, distributed, and hybrid approaches while using different communication
technologies. However, the use of heterogeneous communications architecture for EVs charging can also be used to enhance the
performance of EVs charging approaches. EVs charging with heterogeneous communication architecture has been discussed in
Malhotra et al. (2017). We consider that it is the limitation of our work that we have not explored in more detail the heterogeneous
EVs charging with different optimization approaches.
Combination of all the traditional EVs charging (centralized, distributed, and hybrid) with heterogeneous communication ar-
chitecture can be employed as the heterogeneous EVs charging approach. Heterogeneous EVs provide more flexbility in optimally
charging the EVs. Different limitations of above mentioned traditional EVs charging approaches can also be addressed with the help
of heterogeneous EVs charging.

6. Conclusion

In this paper, we have provided a comprehensive review of optimization approaches, optimization objectives, and various ap-
proaches used for electric vehicles (EVs) charging. Usually, EVs are charged using the centralized, distributed and hybrid approaches
while employing various optimization approaches. Different objectives achieved during this optimal charging of EVs have also been
discussed with respect to various optimization approaches. Various limitations introduced by different optimization approaches

411
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

(centralized, distributed, and hybrid approaches) have also been discussed in conjunction with different optimization approaches.
Renewable energy source integration, power losses, electricity cost, aggregator profit, frequency and voltage regulation, dis-
tribution network issues, and peak load management have been discussed while considering various optimization approaches during
EVs charging. The optimal charging of EVs using various optimization approaches enhance the overall performance of the dis-
tribution grid compared to an un-optimized EVs charging.

References

Abe, Masanori, Saitoh, Hiroumi, 2014. An optimal scheduling of battery output and load control for loss minimizing islanding operation of distribution networks in
emergency. In: IEEE Power Systems Computation Conference (PSCC), 2014, pp. 1–7.
Acha, S., Green, T.C., Shah, N., 2010. Effects of optimised plug-in hybrid vehicle charging strategies on electric distribution network losses. In: IEEE Transmission and
Distribution Conference and Exposition, 2010 PES, pp. 1–6.
Ahmad, Aqueel, Alam, Mohammad Saad, Chabaan, Rakan, 2017. A comprehensive review of wireless charging technologies for electric vehicles ab1. IEEE Trans.
Transp. Electrif.
Ahn, C., Li, C.-T., Peng, H., 2011. Optimal decentralized charging control algorithm for electrified vehicles connected to smart grid. J. Power Sources 196,
10369–10379.
Amjad, Muhammad, Sharif, Muhammad, Afzal, Muhammad Khalil, Kim, Sung Won, 2016. TinyOS-new trends, comparative views, and supported sensing applications:
a review. IEEE Sens. J. 16 (9), 2865–2889.
Amjad, Muhammad, Afzal, Muhammad Khalil, Umer, Tariq, Kim, Byung-Seo, 2017a. QoS-aware and heterogeneously clustered routing protocol for wireless sensor
networks. IEEE Access 5, 10250–10262.
Amjad, Muhammad, Akhtar, Fayaz, Rehmani, Mubashir Husain, Reisslein, Martin, Umer, Tariq, 2017b. Full-duplex communication in cognitive radio networks: a
survey. IEEE Commun. Surv. Tutorials.
Amjad, M., Rehmani, M.H., Mao, S., 2018. Wireless multimedia cognitive radio networks: a comprehensive survey. In: IEEE Commun. Surv. Tutorials.
Amjad, Muhammad, Rehmani, Mubashir Husain, Umer, Tariq, 2018. Electric vehicle scheduling and charging in the smart cities. In: Mouftah, H., Kantarci, M.E.,
Rehmani, M.H. (Eds.) Transportation and Power Grid in Smart Cities: Communication Networks and Services (in press).
An, Jing, Huang, BingYao, Kang, Qi, Zhou, MengChu, 2014. PSO-based method to find electric vehicle’s optimal charging schedule under dynamic electricity price. In:
Networking, Sensing and Control (ICNSC), 2013 10th IEEE International Conference on, pp. 913–918.
Ardakanian, O., Rosenberg, C., Keshav, S., 2013. Distributed control of electric vehicle charging. In: Proceedings of the Fourth International Conference on Future
Energy Systems. ACM, pp. 101–112.
Arias, Andrés, Granada, Mauricio, Castro, Carlos A., 2017. Optimal probabilistic charging of electric vehicles in distribution systems. IET Electr. Syst. Transp.
Arikan, Abdurrahman, Jin, Ruofan, Wang, Bing, Han, Song, Pattipati, Krishna R., Yi, Ping, Zhu, Ting, 2015. Optimal renewable energy transfer via electrical vehicles.
In: IEEE Power and Energy Society Innovative Smart Grid Technologies Conference (ISGT), 2015, pp. 1–5.
Awasthi, Abhishek, Chandra, Dinesh, Rajasekar, S., Singh, Asheesh K., Muruga Perumal, K., 2016. Optimal infrastructure planning of electric vehicle charging stations
using hybrid optimization algorithm. In: Power Systems Conference (NPSC), 2016 National. IEEE, pp. 1–6.
Bai, Xiaoqing, Qiao, Wei, 2015. Robust optimization for bidirectional dispatch coordination of large-scale V2G. IEEE Trans. Smart Grid 1944–1954.
Balram, Pavan, Le, Anh Tuan, Tjernberg, Lina Bertling, 2013. Stochastic programming based model of an electricity retailer considering uncertainty associated with
electric vehicle charging. In: IEEE 10th International Conference on the European Energy Market (EEM), 2013, pp. 1–8.
Banos, Raul, Manzano-Agugliaro, Francisco, Montoya, F.G., Gil, Consolacion, Alcayde, Alfredo, Gómez, Julio, 2011. Optimization methods applied to renewable and
sustainable energy: a review. Renew. Sustain. Energy Rev. 15 (4), 1753–1766.
Bashash, Saeid, Fathy, Hosam K., 2013. Optimizing demand response of plug-in hybrid electric vehicles using quadratic programming. In: American Control
Conference (ACC), 2013, pp. 716–721.
Bashash, Saeid, Fathy, Hosam K., 2014. Cost-optimal charging of plug-in hybrid electric vehicles under time-varying electricity price signals. IEEE Trans. Intell. Transp.
Syst. 15 (5), 1958–1968.
Bashash, Saeid, Moura, Scott J., Forman, Joel C., Fathy, Hosam K., 2011. Plug-in hybrid electric vehicle charge pattern optimization for energy cost and battery
longevity. J. Power Sources 196 (1), 541–549.
Beaude, O., He, Y., Hennebel, M., 2013. Introducing decentralized EV charging coordination for the voltage regulation. In: IEEE/PES ISGT Europe 2013, pp. 1–5.
Bessa, R.J., Matos, M.A., 2012. Economic and technical management of anaggregation agent for electric vehicles:a literature survey. Eur. Trans. Electr. Power 22,
334–350.
Bessa, Ricardo J., Matos, Manuel, 2013. Optimization models for EV aggregator participation in a manual reserve market. IEEE Trans. Power Syst. 28 (3), 3085–3095.
Bessa, R., Soares, F., Lopes, J.A.P., Matos, M.A., 2011. Models for the EV aggregation agent business. In: 2011 IEEE Trondheim PowerTech. IEEE, pp. 1–8.
Bessa, Ricardo J., Matos, Manuel, Soares, Filipe Joel, Peças Lopes, João A., 2012. Optimized bidding of a EV aggregation agent in the electricity market. IEEE Trans.
Smart Grid 3 (1), 443–452.
Brembeck, Jonathan, Ritzer, Peter, 2012. Energy optimal control of an over actuated Robotic Electric Vehicle using enhanced control allocation approaches. In:
Intelligent Vehicles Symposium (IV). IEEE, pp. 322–327.
Brenna, Morris, Foiadelli, Federica, Longo, Michela, Zaninelli, Dario, 2017. A study of the integration of Distributed Generation and EVs in smart cities. In: Electrical
and Electronic Technologies for Automotive, 2017 International Conference of. IEEE, pp. 1–6.
Cai, H., Du, Wenjuan, Yu, Xiao Peng, Gao, Smith, Littler, Tim, Wang, H.F., 2013. Day-ahead optimal charging/discharging scheduling for electric vehicles in micro-
grids, pp. 2–69.
Cao, Yijia, Tang, Shengwei, Li, Canbing, Zhang, Peng, Tan, Yi, Zhang, Zhikun, Li, Junxiong, 2012. An optimized EV charging model considering TOU price and SOC
curve. IEEE Trans. Smart Grid 3 (1), 388–393.
Cao, Yue, Wang, Ning, Kamel, George, Kim, Young-Jin, 2015. An electric vehicle charging management scheme based on publish/subscribe communication frame-
work. IEEE Syst. J.
Celli, G., Ghiani, E., Pilo, F., Pisano, G., Soma, G.G., 2012. Particle swarm optimization for minimizing the burden of electric vehicles in active distribution networks.
In: Power and Energy Society General Meeting, 2012 IEEE, pp. 1–7.
Chammas, M., Chiche, A., Fournie, L., Fidalgo, Jose Nuno, Couto, Mario J., 2013. A multi-scale optimization model to assess the benefits of a smart charging policy for
electrical vehicles. In: PowerTech (POWERTECH), 2013 IEEE Grenoble, pp. 1–6.
Chan, Chieh-Min, Lu, Chan-Nan, Lo, Yuan Liang, 2012. Optimal use of existing distribution feeders to accommodate transportation electrification. In: IEEE
Transactions on Intelligent Transportation Systems, pp. 1941–1950.
Chen, Niangjun, Quek, Tony Q.S., Tan, Chee Wei, 2012. Optimal charging of electric vehicles in smart grid: characterization and valley-filling algorithms. In: IEEE
Third International Conference on Smart Grid Communications (SmartGridComm), 2012, pp. 13–18.
Cheng, Lin, Chang, Yao, Huang, Renle, 2015. Mitigating Voltage Problem in Distribution System With Distributed Solar Generation Using Electric Vehicles.
Clement, Kristien, Haesen, Edwin, Driesen, Johan, 2009. Coordinated charging of multiple plug-in hybrid electric vehicles in residential distribution grids. In: IEEE/
PES Power Systems Conference and Exposition, 2009. PSCE’, pp. 1–7.
Clement, Kristien, Haesen, Edwin, Driesen, Johan, 2009b. Coordinated charging of multiple plug-in hybrid electric vehicles in residential distribution grids. In: IEEE/
PES Power Systems Conference and Exposition. IEEE, pp. 1–7.
Clement-Nyns, K., Haesen, E., Driesen, J., 2010. The impact of charging plug-in hybrid electric vehicles on a residential distribution grid. IEEE Trans. Power Syst. 25,
371–380.

412
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Conway, Thomas, 2017. On the effects of a routing and reservation system on the electric vehicle public charging network. IEEE Trans. Intell. Transp. Syst.
Costa, Inaldo C., Rosa, M., Carvalho, Luis, Bremermann, Leonardo, Iria, J., 2014. Identifying benefits between the integration of Electric Vehicles and renewable power
usage. In: IEEE 8th International Power Engineering and Optimization Conference (PEOCO), 2014, pp. 341–346.
Dallinger, D., Wietschel, M., 2012. Grid integration of intermittent renewable energy sources using price-responsive plug-in electric vehicles. Renew. Sustain. Energy
Rev. 16, 3370–3382.
Deilami, S., Member, S., Masoum, A.S., Moses, P.S., Masoum, M.A.S., Member, S., 2011. Real- time coordination of plug-in electric vehicle charging in smart grids to
minimize power losses and improve voltage profile. IEEE Trans. Smart Grid 2, 456–467.
Desai, Chirag, Berthold, Florence, Williamson, Sheldon S., 2010. Optimal drivetrain component sizing for a plug-in hybrid electric transit bus using multi-objective
genetic algorithm. In: Electric Power and Energy Conference (EPEC). IEEE, pp. 1–5.
Deshpande, Ajay, Murali, Pramod, 2013. Pricing long-term permits and scheduling of electric vehicle charging in parking lots with shared resources. In: Control
Conference (ECC), 2013 European, pp. 3584–3589.
Di Giorgio, A., Liberati, F., Canale, S., 2014. Electric vehicles charging control in a smart grid: a model predictive control approach. Control Eng. Pract. 22, 147–162.
Dixit, Mukul, Roy, Ranjit, 2015. PSO-CFA based optimal placement of EVs in radial distribution network for loss minimization. In: IEEE International Conference on
Electrical, Computer and Communication Technologies (ICECCT), 2015, pp. 1–5.
Donadee, Jonathan, Ilić, Marija, 2012. Stochastic co-optimization of charging and frequency regulation by electric vehicles. In: North American Power Symposium
(NAPS). IEEE, pp. 1–6.
Dutta, Promiti, Boulanger, Albert, 2014. Game theoretic approach to offering participation incentives for electric vehicle-to-vehicle charge sharing. In: Transportation
Electrification Conference and Expo (ITEC). IEEE, pp. 1–5.
Fan, Z., 2012. A distributed demand response algorithm and its application to PHEV charging in smart grids. IEEE Trans. Smart Grid 3, 1280–1290.
Fang, Xi, Misra, Satyajayant, Xue, Guoliang, Yang, Dejun, 2012. Smart grid-the new and improved power grid: a survey. IEEE Commun. Surv. Tutorials 14 (4),
944–980.
Fazeli, Amir, Mark Johnson, C., Sumner, Mark, Christopher, Edward, 2015. A novel stochastic modelling approach for electric vehicle charging power and energy
requirements. In: Innovative Smart Grid Technologies Conference (ISGT), 2015 IEEE Power and Energy Society. IEEE, pp. 1–5.
Feng, Bowei, Yu, Rong, Lai, Yue, 2013. Efficient and fair scheduling of PHEV charging with neuro dynamic programming. In: Communications and Networking in
China (CHINACOM), 2013 8th International ICST Conference on, pp. 930–935.
Freire, Rui, Delgado, Joaquim, Santos, J., Almeida, A., 2010. Integration of renewable energy generation with EV charging strategies to optimize grid load balancing.
In: IEEE Annual Conference on Intelligent Transportation Systems, pp. 392–396.
Galus, M.D., Andersson, G., 2008. Demand management of grid connected plug-in hybrid electric vehicles (PHEV). In: IEEE Energy 2030 Conference, 2008, pp. 1–8.
Gan, L., Topcu, U., Low, S., 2011. Optimal decentralized protocol for electric vehicle charging. In: IEEE 50th Conference on Decision and Control and European Control
Conference (CDC-ECC), Dec. 2011, pp. 5798–5804.
Gan, L., Topcu, U., Low, S.H., 2012. Optimal decentralized protocol for electric vehicle charging. IEEE Trans. Power Syst. 28, 1–12.
García-Villalobos, J., Zamora, I., San Martín, J.I., Asensio, F.J., Aperribay, V., 2014. Plug-in electric vehicles in electric distribution networks: a review of smart
charging approaches. Renew. Sustain. Energy Rev. 38, 717–731.
Garey, M.R., Johnson, D.S., 1979. Computers and Intractability: A Guide to the Theory of NP-completeness. W.H. Freeman and Company, San Francisco.
Gerding, E.H., Robu, V., Stein, S., Parkes, D.C., Rogers, A., Jennings, N.R., 2011. Online mechanism design for electric vehicle charging. In: The 10th International
Conference on Autonomous Agents and Multiagent Systems – 2011, pp. 811–818.
Gonzalez Vaya, M., Andersson, G., 2012. Centralized and decentralized approaches to smart charging of plug-in vehicles. In: IEEE Power Energy Society General
Meeting, 2012, pp. 1–8.
Grau Unda, I., Papadopoulos, P., Skarvelis-Kazakos, S., Cipcigan, L.M., Jenkins, N., Zabala, E., 2014. Management of electric vehicle battery charging in distribution
networks with multi-agent systems. Electr. Power Syst. Res. 110, 172–179.
Guille, C., Gross, G., Design of a conceptual framework for the V2G implementation. In: IEEE Energy 2030 Conference: Atlanta, GA, 2008, pp. 1–3.
Gusrialdi, Azwirman, Qu, Zhihua, Simaan, Marwan A., 2017. Distributed scheduling and cooperative control for charging of electric vehicles at highway service
stations. IEEE Trans. Intell. Transp. Syst.
Habib, Salman, Kamran, Muhammad, Rashid, Umar, 2015. Impact analysis of vehicle-to-grid technology and charging strategies of electric vehicles on distribution
networks-a review. J. Power Sources 277, 205–214.
Hadi Amini, M., Kargarian, Amin, Karabasoglu, Orkun, 2016. ARIMA-based decoupled time series forecasting of electric vehicle charging demand for stochastic power
system operation. Electr. Power Syst. Res. 140, 378–390.
Hadley, Stanton W., Tsvetkova, Alexandra A., 2009. Potential impacts of plug-in hybrid electric vehicles on regional power generation. Electr. J. 22 (10), 56–68.
Han, S., Han, S., Sezaki, K., 2010a. Development of an optimal vehicle-to-grid aggregator for frequency regulation. IEEE Trans. Smart Grid 1 (1), 65–72.
Han, S., Member, S., Han, S., Sezaki, K., 2010b. Development of an optimal vehicle-to-grid aggregator for frequency regulation. IEEE Trans. Smart Grid 1, 65–72.
Han, Sekyung, Han, Soohee, Sezaki, Kaoru, 2011. Optimal control of the plug-in electric vehicles for V2G frequency regulation using quadratic programming. In:
Innovative Smart Grid Technologies (ISGT), 2011 IEEE PES, pp. 1–6.
He, Y., Venkatesh, B., Guan, L., 2012. Optimal scheduling for charging and discharging of electric vehicles. IEEE Trans. Smart Grid 3 (3), 1095–1105.
Herrera, Victor Isaac, Saez-de-Ibarra, Andoni, Milo, Aitor, Gaztanaga, Haizea, Camblong, Haritza, 2015. Optimal energy management of a hybrid electric bus with a
battery-supercapacitor storage system using genetic algorithm. In: Electrical Systems for Aircraft, Railway, Ship Propulsion and Road Vehicles (ESARS), 2015
International Conference on, pp. 1–6.
Huang, Bufu, Wang, Zhancheng, Xu, Yangsheng, 2006. Multi-objective genetic algorithm for hybrid electric vehicle parameter optimization. In: Intelligent Robots and
Systems, 2006 IEEE/RSJ International Conference on. IEEE, pp. 5177–5182.
Huang, J., Gupta, V., Huang, Y.-F., Scheduling algorithms for PHEV charging in shared parking lots. In: IEEE American Control Conference (ACC), 2012, pp. 276–281.
Huang, Shaojun, Pillai, Jayakrishnan Radhakrishna, Bak-Jensen, Brigitte, Thogersen, Paul, 2013. Voltage support from electric vehicles in distribution grid. In: 2013
15th European Conference on Power Electronics and Applications (EPE), pp. 1–8.
Huang, Qilong, Jia, Qing-Shan, Xia, Li, Guan, Xiaohong, 2014a. Event-based optimization for stochastic matching EV charging load with uncertain renewable energy.
In: Intelligent Control and Automation (WCICA), 2014 11th World Congress on. IEEE, pp. 794–799.
Huang, Shanjin, Wu, Qingyao, Oren, Shmuel S., Li, Ruodai, Liu, Zhe, 2014. Distribution locational marginal pricing through quadratic programming for congestion
management in distribution networks.
Huang, Qilong, Jia, Qing-Shan, Guan, Xiaohong, 2015. Multi-timescale optimization between distributed wind generators and electric vehicles in microgrid. In: IEEE
International Conference on Automation Science and Engineering (CASE), 2015, pp. 671–676.
Zechun, Hu, Song, Yonghua, Xu, Zhiwei, Luo, Zhuowei, Zhan, Kaiqiao, Jia, Long, 2012. Impacts and utilization of electric vehicles integration into power systems. In:
Dianji, Zhongguo, Xuebao, Gongcheng (Eds.), Proceedings of the Chinese Society of Electrical Engineering, vol. 32, no. 4. Chinese Society for Electrical
Engineering, pp. 1-10.
Hu, J., You, S., Lind, M., Ostergaard, J., 2014. Coordinated charging of electric vehicles for congestion prevention in the distribution grid. IEEE Trans. Smart Grid 5 (2),
703–711.
Hutson, C., Venayagamoorthy, G.K., Corzine, K.A., 2008. Intelligent scheduling of hybrid and electric vehicle storage capacity in a parking lot for profit maximization
in grid power transactions. In: IEEE Energy 2030 Conference, 2008, pp. 1–8.
Iversen, E.B., Morales, J.M., Madsen, H., 2014. Optimal charging of an electric vehicle using a Markov decision process. Appl. Energy 123, 1–12.
Jain, Manu, Desai, Chirag, Kharma, Nawwaf, Williamson, Sheldon S., 2009. Optimal powertrain component sizing of a fuel cell plug-in hybrid electric vehicle using
multi-objective genetic algorithm. In: Industrial Electronics, 2009. IECON’09. 35th Annual Conference of IEEE, pp. 3741–3746.
Jamil, F., Javaid, A., Umer, T., Rehmani, M.H., 2017. A comprehensive survey of network coding in vehicular ad-hoc networks. Wireless Netw. 23 (8), 2395–2414.
Jiang B., Fei, Y., 2013. Decentralized scheduling of PEV on-street parking and charging for smart grid reactive power compensation. In: IEEE Innovative Smart Grid

413
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Technologies (ISGT), 2013, Feb. 2013, pp. 1–6.

Jiang, Zhi-ling, Wei-Rong, Chen, Chao-hua, Dai, Zhan-li, Cheng, 2010. Optimal energy management for a fuel cell hybrid locomotive. In: Fuzzy Systems and
Knowledge Discovery (FSKD), 2010 Seventh International Conference on, vol. 3, pp. 1320–1323.
Jiménez, Alejandra, García, Norberto, 2012. Voltage unbalance analysis of distribution systems using a three-phase power flow ans a Genetic Algorithm for PEV fleets
scheduling. In: IEEE Power and Energy Society General Meeting, 2012, pp. 1–8.
Jin, Chenrui, Mojdehi, Mohammad Nikkhah, Ghosh, Prosenjit, 2012. A methodology to design a stochastic cost efficient DER scheduling considering environmental
impact. In: IEEE International Conference on Smart Grid Engineering (SGE), 2012, pp. 1–6.
Jin, Chenrui, Tang, Jian, Ghosh, Prosenjit, 2013a. Optimizing electric vehicle charging with energy storage in the electricity market. IEEE Trans. Smart Grid 4 (1),
311–320.
Jin, Chenrui, Sheng, Xiang, Ghosh, Prosenjit, 2013. Energy efficient algorithms for electric vehicle charging with intermittent renewable energy sources. In: IEEE
Power and Energy Society General Meeting (PES), 2013, pp. 1–5.
Jin, Chenrui, Tang, Jian, Ghosh, Prosenjit, 2013. Optimizing electric vehicle charging with energy storage in the electricity market. IEEE Trans. Smart Grid 4 (1),
311–320.
Jin, Chenrui, Sheng, Xiang, Ghosh, Prosenjit, 2014. Optimized electric vehicle charging with intermittent renewable energy sources. IEEE J. Sel. Top. Signal Process. 8
(6), 1063–1072.
Kang, J., Duncan, S.J., Mavris, D.N., 2013. Real-time scheduling techniques for electric vehicle charging in support of frequency regulation. Procedia Comput. Sci. 16,
767–775.
Karfopoulos, E.L., Papadopoulos, P., Skarvelis-Kazakos, S., Grau, I., Cipcigan, L.M., Hatziargyriou, N., et al., 2011. Introducing electric vehicles in the microgrids
concept. In: 2011 16th International Conference Intelligent Systems Applications to Power Systems. Hersonissos, pp. 1–6.
Kaschub, T., Jochem, P., Fichtner, W., 2016. Solar energy storage in German households: profitability, load changes, and flexibility. Energy Policy 98, 520–532.
Kazerooni, Maryam, Kar, Narayan C., 2012. Optimal load management of EV battery charging and optimization of harmonic impacts on distribution transformers. In:
25th IEEE Canadian Conference on Electrical & Computer Engineering (CCECE), 2012, pp. 1–4.
Kempton, W., Tomić, J., 2005a. Vehicle-to-grid power implementation: from stabilizing the grid to supporting large-scale renewable energy. J. Power Sources 144,
280–294.
Kempton, Willett, Tomić, Jasna, 2005b. Vehicle-to-grid power fundamentals: calculating capacity and net revenue. J. Power Sources 144 (1), 268–279.
Ko, Young Dae, Jang, Young Jae, 2013. The optimal system design of the online electric vehicle utilizing wireless power transmission technology. IEEE Trans. Intell.
Transp. Syst. 14 (3), 1255–1265.
Ko, Kab Seok, Sung, Dan Keun. A centralized EV charging scheme based on user satisfaction fairness and cost. In: IEEE Innovative Smart Grid Technologies-Asia (ISGT
Asia), 2013, pp. 1–4.
Lee, Hyeoun-Dong, Sul, Seung-Ki, 1998. Fuzzy-logic-based torque control strategy for parallel-type hybrid electric vehicle. IEEE Trans. Industr. Electron. 45 (4),
625–632.
Le, Tan N., Al-Rubaye, Saba, Liang, Hao, Jun Choi, Bong, 2015. Dynamic charging and discharging for electric vehicles in microgrids. In: IEEE International Conference
on Communication Workshop (ICCW), 2015, pp. 2018–2022.
Li, Xiaodong, Chen, Yan, Wang, Junmin, 2012. In-wheel motor electric ground vehicle energy management strategy for maximizing the travel distance. In: American
Control Conference (ACC), 2012, pp. 4993–4998.
Li, Hongyang, Dán, György, Nahrstedt, Klara, 2017. Portunes+: privacy-preserving fast authentication for dynamic electric vehicle charging. IEEE Trans. Smart Grid.
Lin, Junhao, Leung, Ka-Cheong, Li, Victor OK, 2014. Optimal scheduling with vehicle-to-grid regulation service. IEEE Internet Things J. 1 (6), 556–569.
Liu, H., Hu, Z., Song, Y., Lin, J., 2013a. Decentralized vehicle-to-grid control for primary frequency regulation considering charging demands. IEEE Trans. Power Syst.
28, 3480–3489.
Liu, Zhipeng, Wen, Fushuan, Ledwich, Gerard, 2013b. Optimal planning of electric-vehicle charging stations in distribution systems. IEEE Trans. Power Delivery 28
(1), 102–110.
Liu, Hong, Guo, Jianyi, Zeng, Pingliang, 2014. A controlled electric vehicle charging strategy considering regional wind and PV. In: IEEE/PES General Meeting—
Conference and Exposition, 2014, pp. 1–5.
Lopes, J.A.P., Hatziargyriou, N., Mutale, J., Djapic, P., Jenkins, N., 2007. Integrating distributed generation into electric power systems: a review of drivers, challenges
and opportunities. Electr. Power Syst. Res. 77, 1189–1203.
Lopes, J.A.P., Rocha Almeida, P.M., Soares, F.J., 2009. Using vehicle-to-grid to maximize the integration of intermittent renewable energy resources in islanded
electric grids. In: IEEE International Conference on Clean Electrical Power.: Capri, pp. 290–295.
Lopes, J.A.P., Soares, F.J., Almeida, P.M.R., 2011. Integration of electric vehicles in the electric power system. Proc. IEEE 99, 168–183.
Lu, Xinyi, Liu, Nian, Chen, Qifang, Zhang, Jianhua, 2014. Multi-objective optimal scheduling of a DC micro-grid consisted of PV system and EV charging station. In:
IEEE Innovative Smart Grid Technologies-Asia (ISGT Asia), 2014, pp. 487–491.
Lu, Fei, Zhang, Hua, Mi, Chris, 2018. A two-plate capacitive wireless power transfer system for electric vehicle charging applications. IEEE Trans. Power Electron. 33
(2), 964–969.
Luo, X., Xia, S., Chan, K.W., 2014a. A decentralized charging control strategy for plug-in electric vehicles to mitigate wind farm intermittency and enhance frequency
regulation. J. Power Sources 248, 604–614.
Luo, Chaomin, Wu, Yu-Ting, Krishnan, Mohan, Paulik, Mark, Jan, Gene Eu, Gao, Jiyong, 2014. An effective search and navigation model to an auto-recharging
charging station of driverless vehicles. In: Computational Intelligence in Vehicles and Transportation Systems (CIVTS), 2014 IEEE Symposium on, pp. 100–107.
Ma, Z., Callaway, D., Hiskens, I., 2010. Decentralized charging control for large populations of plug-in electric vehicles: application of the Nash certainty equivalence
principle. In: IEEE International Conference on Control Applications (CCA), 2010, pp. 191–195.
Ma, Wann-Jiun, Gupta, Vijay, Topcu, Ufuk, 2016. Distributed charging control of electric vehicles using online learning. IEEE Trans. Autom. Control.
Malhotra, Akshay, Binetti, Giulio, Davoudi, Ali, Schizas, Ioannis D., 2017. Distributed power profile tracking for heterogeneous charging of electric vehicles. IEEE
Trans. Smart Grid.
Mao, Tian, Lau, W.H., Shum, Chong, Chung, Henry, Tsang, K.F., Tse, Norman C.F., 2014. An intelligent scatter search (ISS) algorithm for scheduling of charging a
single EV. In: IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC), 2014, pp. 1–5.
Mauri, Giuseppe, Valsecchi, Antonio, 2012. The role of fast charging stations for electric vehicles in the integration and optimization of distribution grid with
renewable energy sources 227–227.
Mayr, Christian H., Fleck, Andreas, Jakubek, Stefan, 2011. Hybrid powertrain control using optimization and cycle based predictive control algorithms. In: Control and
Automation (ICCA), 2011 9th IEEE International Conference on, pp. 937–944.
McArthur, S.D.J., Davidson, E.M., Catterson, V.M., Dimeas, A.L., Hatziargyriou, N.D., Ponci, F., et al., 2007. Multi-agent systems for power engineering applications-
Part I: concepts, approaches, and technical challenges. IEEE Trans. Power Syst. 22, 1743–1752.
Mehar, Sara, Mohammed Senouci, Sidi 2013. An optimization location scheme for electric charging stations. In: Smart Communications in Network Technologies
(SaCoNeT), 2013 International Conference on, vol. 1, pp. 1–5.
Mehboob, Nafeesa, Canizares, Claudio, Rosenberg, Catherine, 2014. Day-ahead dispatch of PEV loads in a residential distribution system. In: PES General Meeting—
Conference & Exposition, 2014 IEEE, pp. 1–5.
Mets, K., Verschueren, T., Haerick, W., Develder, C., De Turck, F., 2010. Optimizing smart energy control strategies for plug-in hybrid electric vehicle charging. In:
IEEE/IFIP Network Operations and Management Symposium Workshops (NOMS Wksps), 2010, pp. 293–299.
Mets, Kevin, D’hulst, Reinhilde, Develder, Chris, 2012. Comparison of intelligent charging algorithms for electric vehicles to reduce peak load and demand variability
in a distribution grid. J. Commun. Netw. 14 (6), 672–681.
Mierau, Michael, Kohrs, Robert, Wittwer, Christof, 2012. A distributed approach to the integration of electric vehicles into future smart grids. In: 3rd IEEE PES
International Conference and Exhibition on Innovative Smart Grid Technologies (ISGT Europe), 2012, pp. 1–7.

414
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Mitsukuri, Yuki, Hara, Ryoichi, Kita, Hajime, Kamiya, Eiji, Kataoka, Yasuyuki, Taki, Shota, Kogure, Eiji, 2012. Validation of voltage regulation method in distribution
system utilizing electric vehicles. In: 2012 3rd IEEE PES International Conference and Exhibition on Innovative Smart Grid Technologies (ISGT Europe). IEEE,
pp. 1–7.
Mojdehi, Mohammad Nikkhah, Ghosh, Prasanta, Fardad, Makan, 2015. Energy and cost minimization of bidirectional frequency regulation service by EV following
FERC order 755. In: Power and Energy Society General Meeting, 2015. IEEE, pp. 1–5.
Moses, Paul S., Deilami, Sara, Masoum, Amir S., Masoum, Mohammad A.S., 2010. Power quality of smart grids with plug-in electric vehicles considering battery
charging profile. In: IEEE/PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe), 2010, pp. 1–7.
Mouftah, Hussein T., Erol-Kantarci, Melike, Rehmani, Mubashir Husain, 2017. Transportation and Power Grid in Smart Cities: Communication Networks and Services.
John Wiley, UK.
Mozafar, Mostafa Rezaei, Moradi, Mohammad H., Amini, M. Hadi, 2017. A Simultaneous Approach for Optimal Allocation of Renewable Energy Sources and Electric
Vehicle Charging Stations in Smart Grids based on Improved GA-PSO Algorithm, vol. 32, pp. 627637.
Mullen, Sara Kathryn, 2009. Plug-in Hybrid Electric Vehicles as a Source of Distributed Frequency Regulation (PhD diss.). University of Minnesota.
Muñoz, Enrique, Ruspini, Enrique, 2012. Using fuzzy queuing theory to analyze the impact of electric vehicles on power networks. In: Fuzzy Systems (FUZZ-IEEE),
2012 IEEE International Conference on, pp. 1–8.
Murgovski, Nikolce, Johannesson, Lars Mardh, Sjoberg, Jonas, 2013. Engine on/off control for dimensioning hybrid electric powertrains via convex optimization. IEEE
Trans. Veh. Technol. 62 (7), 2949–2962.
Neaimeh, Myriam, Hill, Graeme, Hubner, Yvonne, Blythe, Phil T., 2013. Routing systems to extend the driving range of electric vehicles. Intell. Transp. Syst. IET 7 (3),
327–336.
Nguyen, H.K., Bin, Song J., 2012. Optimal charging and discharging for multiple PHEVs with demand side management in vehicle-to-building. J. Commun. Netw. 14,
662–671.
Nguyen, D.T., Le, L.B., 2014. Joint optimization of electric vehicle and home energy scheduling considering user comfort preference. IEEE Trans. Smart Grid 5 (1),
188–199.
Nguyen, Van-Linh, Tran-Quoc, Tuan, Bacha, Seddik, Nguyen, Be, 2014. Charging strategies to minimize the peak load for an electric vehicle fleet. In: Industrial
Electronics Society, IECON 2014-40th Annual Conference of the IEEE, pp. 3522–3528.
Nguyen, Hoang, Zhang, Cishen, Mahmud, Md., (2015). Optimal Coordination of G2V and V2G to Support Power Grids with High Penetration of Renewable Energy.
Oliveira, D.Q., Zambroni de Souza a, C., Delboni, L.F.N., 2013. Optimal plug-in hybrid electric vehicles recharge in distribution power systems. Electr. Power Syst. Res.
98, 77–85.
Pardalos, P.M., Resende, M.G.C., 2002. Handbook of Applied Optimization. Oxford University Press.
Pearl, Judea, 1984. Heuristics: intelligent search strategies for computer problem solving.
Pillai, J.R., Bak-Jensen, B., 2010. Vehicle-to-grid for islanded power system operation in Bornholm. In: IEEE/PES General Meeting, pp. 1–8.
Pillai, J.R., Bak-Jensen, B., 2010. Vehicle-to-grid systems for frequency regulation in an Islanded Danish distribution network. In: IEEE Vehicle Power and Propulsion
Conference. Lille, pp. 1–6.
Pourabdollah, Mitra, Murgovski, Nikolce, Grauers, Anders, Egardt, Bo, 2013. Optimal sizing of a parallel PHEV powertrain. IEEE Trans. Veh. Technol. 62 (6),
2469–2480.
Pourazarm, Sepideh, Cassandras, Christos G., Malikopoulos, Andreas, 2014. Optimal routing of electric vehicles in networks with charging nodes: a dynamic pro-
gramming approach. In: Electric Vehicle Conference (IEVC), 2014 IEEE International. IEEE, pp. 1–7.
Qian, Kejun, Zhou, Chengke, Allan, Malcolm, Yuan, Yue, 2011. Modeling of load demand due to EV battery charging in distribution systems. IEEE Trans. Power Syst.
26 (2), 802–810.
Qi, W., Xu, Z., Shen, Z.J.M., Hu, Z., Song, Y., 2014. Hierarchical coordinated control of plug-in electric vehicles charging in multifamily dwellings. IEEE Trans. Smart
Grid 5 (3), 1465–1474.
Rahimi, S., Zhu, K., Massucco, S., Silvestro, F., Steen, D., 2014. Using the advanced DMS functions to handle the impact of plug-in electric vehicles on distribution
networks. In: IEEE International Electric Vehicle Conference (IEVC), 2014, pp. 1–8.
Rahman, Imran, Vasant, Pandian, Singh, Balbir Singh Mahinder, Abdullah-Al-Wadud, M., 2015. Swarm intelligence-based optimization for PHEV charging stations.
Handbook Res. Swarm Intell. Eng. 374.
Rahman, Imran, Vasant, Pandian M., Singh, Balbir Singh Mahinder, Abdullah-Al-Wadud, M., Adnan, Nadia, 2016a. Review of recent trends in optimization techniques
for plug-in hybrid, and electric vehicle charging infrastructures. Renew. Sustain. Energy Rev. 58, 1039–1047.
Rahman, Imran, Vasant, Pandian M., Singh, Balbir Singh Mahinder, Abdullah-Al-Wadud, M., 2016b. On the performance of accelerated particle swarm optimization
for charging plug-in hybrid electric vehicles. Alexandria Eng. J. 55 (1), 419–426.
Ratnam, Elizabeth L., Weller, Steven R., Kellett, Christopher M., 2013. An optimization-based approach for assessing the benefits of residential battery storage in
conjunction with solar PV. In: Bulk Power System Dynamics and Control-IX Optimization, Security and Control of the Emerging Power Grid (IREP), 2013 IREP
Symposium, pp. 1–8.
Rehan, Maaz, Hasbullah, Halabi, Faye, Ibrahima, Rehan, Waqas, Chughtai, Omer, Rehmani, Mubashir Husain, 2017. ZGLS: a novel flat quorum-based and reliable
location management protocol for VANETs. Wireless Netw. 1–19.
Rezaei, P., Frolik, J., Hines, P.D.H., 2014. Packetized plug-in electric vehicle charge management. IEEE Trans. Smart Grid 5 (2), 642–650.
Richardson, D.B., 2013. Electric vehicles and the electric grid: a review of modeling approaches, impacts, and renewable energy integration. Renew. Sustain. Energy
Rev. 19, 247–254.
Richardson, P., Flynn, D., Keane, A., 2012a. Local versus centralized charging strategies for electric vehicles in low voltage distribution systems. IEEE Trans. Smart Grid
3, 1020–1028.
Richardson, P., Flynn, D., Keane, A., 2012b. Optimal charging of electric vehicles in low-voltage distribution systems. IEEE Trans. Power Syst. 27 (1), 268–279.
Rotering, Niklas, Ilic, Marija, 2011. Optimal charge control of plug-in hybrid electric vehicles in deregulated electricity markets. IEEE Trans. Power Syst. 26 (3),
1021–1029.
Ruiqin, Duan, Zhongjing, Ma, 2015. Resident electric vehicles charging optimization strategy in the smart grid. In: IEEE Control Conference (CCC), 2015, pp.
9054–9059.
Saberbari, Erfan, Saboori, Hedayat, 2014. Evaluating PHEV impacts on domestic distribution grid in terms of power losses and voltage drop. In: Electrical Power
Distribution Networks (EPDC), 2014 19th Conference on, pp. 52–58.
Sabillon, Carlos, Dominguez, Ozy Melgar, Franco, John, Lavorato, Marina, Rider, Marcos J., 2017. Volt-VAr control and energy storage device operation to improve
the electric vehicle charging coordination in unbalanced distribution networks. IEEE Trans. Sustain. Energy.
Said, Dhaou, Cherkaoui, Soumaya, Khoukhi, Lyes, 2012. Scheduling protocol with load managementfor EV charging. In: IEEE Global Communications Conference
(GLOBECOM), 2014, pp. 362–367.
Seddig, K., Jochem, P., Fichtner, W., 2017. Integrating renewable energy sources by electric vehicle fleets under uncertainty. Energy 141, 2145–2153.
Shaoyun, Ge, Liu, Liu Hong Huang, 2012. Optimization of peak-valley TOU power price time-period in ordered charging mode of electric vehicle. Power Syst. Prot.
Control 40 (10), 1–5.
Sheikhi, a, Bahrami, S., Ranjbar, A.M., Oraee, H., 2013. Strategic charging method for plugged in hybrid electric vehicles in smart grids; a game theoretic approach.
Int. J. Electr. Power Energy Syst. 53, 499–506.
Siddiqi, Umair Farooq, Shiraishi, Yoichi, Sait, Sadiq M., 2011. Multi-constrained route optimization for electric vehicles (EVs) using particle swarm optimization
(PSO). In: 11th International Conference on Intelligent Systems Design and Applications (ISDA), 2011, pp. 391–396.
Siddiqi, Umair F., Shiraishi, Yoichi, Sait, Sadiq M., 2011b. Multi constrained route optimization for electric vehicles using SimE. In: Soft Computing and Pattern
Recognition (SoCPaR), 2011 International Conference of. IEEE, pp. 376–383.
Singh, Anant Kumar, 2016. Novel bidirectional single-phase single-stage isolated AC-DC converter with PFC for charging of electric vehicles (PhD diss.).

415
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

Singh, Mukesh, Kar, Indrani, Kumar, Praveen, 2010. Influence of EV on grid power quality and optimizing the charging schedule to mitigate voltage imbalance and
reduce power loss. In: 14th IEEE International Power Electronics and Motion Control Conference (EPE/PEMC), 2010, pp. T2-196.
Singh, M., Kumar, P., Kar, I., 2012. Implementation of vehicle to grid infrastructure using fuzzy logic controller. IEEE Trans. Smart Grid 3, 565–577.
Soares, F.J., Almeida, P.M.R., Lopes, J.A.P., 2014. Quasi-real-time management of electric vehicles charging. Electr. Power Syst. Res. 108, 293–303.
Sojoudi, Somayeh, Low, Steven H., 2011. Optimal charging of plug-in hybrid electric vehicles in smart grids. In: Power and Energy Society General Meeting. IEEE,
pp. 1–6.
Sortomme, E., El-Sharkawi, M.A., 2011. Optimal charging strategies for unidirectional vehicle-to-grid. IEEE Trans. Smart Grid 2 (1), 131–138.
Sortomme, Eric, El-Sharkawi, Mohamed, 2012a. Optimal scheduling of vehicle-to-grid energy and ancillary services. IEEE Trans. Smart Grid 3 (1), 351–359.
Sortomme, Eric, El-Sharkawi, Mohamed, 2012b. Optimal combined bidding of vehicle-to-grid ancillary services. IEEE Trans. Smart Grid 3 (1), 70–79.
Sortomme, E., Member, S., Hindi, M.M., Macpherson, S.D.J., Venkata, S.S., Fellow, L., 2011. Coordinated charging of plug-in hybrid electric vehicles to minimize
distribution system losses. IEEE Trans. Smart Grid 2, 198–205.
Subramanian, Ananth, Garcia, M., Dominguez-Garcia, Alejandro, Callaway, Duncan, Poolla, K.Varaiya, Pravin, 2012. Real-time scheduling of deferrable electric loads.
In: American Control Conference (ACC), 2012, pp. 3643–3650.
Su, Wencong, Eichi, Habiballah, Zeng, Wente, Chow, Mo-Yuen, 2012. A survey on the electrification of transportation in a smart grid environment. IEEE Trans. Industr.
Inf. 8 (1), 1–10.
Su, Chun-Lien, Leou, Rong-Ceng, Yang, Jun-Chang, Lu, Chan-Nan, 2013. Optimal electric vehicle charging stations placement in distribution systems. In: 2013-39th
Annual Conference of the IEEE Industrial Electronics Society, pp. 2121–2126.
Sundstrom, O., Binding, C., 2010. Optimization methods to plan the charging of electric vehicle fleets. Int. J. Commun. 1 (2), 45–51.
Sundstrom, O., Binding, C., 2012. Flexible charging optimization for electric vehicles considering distribution grid constraints. IEEE Trans. Smart Grid 3, 26–37.
Sun, Bo, Tan, Xiaoqi, Tsang, Danny H.K., 2014. Optimal charging operation of battery swapping stations with QoS guarantee. In: Smart Grid Communications
(SmartGridComm), 2014 IEEE International Conference on, pp. 13–18.
Sweda, Timothy M., Klabjan, Diego, 2012. Finding minimum-cost paths for electric vehicles. In: Electric Vehicle Conference (IEVC), 2012 IEEE International, pp. 1–4.
Tan, Z., Yang, P., Nehorai, A., 2014. An optimal and distributed demand response strategy with electric vehicles in the smart grid. IEEE Trans. Smart Grid 5 (2),
861–869.
Tan, Kang Miao, Ramachandaramurthy, Vigna K., Yong, Jia Ying, 2016. Integration of electric vehicles in smart grid: a review on vehicle to grid technologies and
optimization techniques. Renew. Sustain. Energy Rev. 53, 720–732.
Tian, Wenqi, He, Jinghan, Jiang, Jiuchun, NIU, Liyong, WANG, Xiaojun, 2012. Multi-objective optimization of charging dispatching for electric vehicle battery
swapping station based on adaptive mutation particle swarm optimization. Power Syst. Technol. 11, 006.
Tokudome, M., Tanaka, K., Senjyu, T., Yona, A., Funabashi, T., Kim, C.-H., 2009. Frequency and voltage control of small power systems by decentralized controllable
loads. In: Int. Conf. Power Electron. Drive Syst. IEEE, Taipei (Taiwan), pp. 666–671.
Torabi, Roham, Gomes, Alvaro, 2014. Optimizing the coordinated charging of a group of electric vehicles. In: IEEE Vehicle Power and Propulsion Conference (VPPC),
2014, pp. 1–6.
Tushar, M.H.K., Assi, C., Maier, M., Uddin, M.F., 2014. Smart microgrids: optimal joint scheduling for electric vehicles and home appliances. IEEE Trans. Smart Grid 5
(1), 239–250.
Ugirumurera, Juliette, Haas, Zygmunt J., 2017. Optimal capacity sizing for completely green charging systems for electric vehicles. IEEE Trans. Transp. Electrification
3 (3), 565–577.
Umer, Tariq, Amjad, Muhammad, Afzal, Muhammad Khalil, Aslam, Muhammad, 2016a. Hybrid rapid response routing approach for delay-sensitive data in hospital
body area sensor network. In: Proceedings of the 7th International Conference on Computing Communication and Networking Technologies. ACM, pp. 3.
Umer, Tariq, Amjad, Muhammad, Shah, Nadir, Ding, Zhiguo, 2016b. Modeling vehicles mobility for connectivity analysis in vanet. In: Intelligent Transportation
Systems. Springer International Publishing, pp. 221–239.
Vagropoulos, Stylianos, Bakirtzis, Anastasios G., 2013. Optimal bidding strategy for electric vehicle aggregators in electricity markets. IEEE Trans. Power Syst. 28 (4),
4031–4041.
Vagropoulos, Stylianos, Kyriazidis, Dimitrios K., Bakirtzis, Anastasios G. (2016). Real-Time Charging Management Framework for Electric Vehicle Aggregators in a
Market Environment.
Valentine, Keenan, Temple, William G., Max Zhang, K., 2011. Intelligent electric vehicle charging: rethinking the valley-fill. J. Power Sources 196 (24), 10717–10726.
Vasant, Pandian M., Rahman, Imran, Singh, Balbir Singh Mahinder, Abdullah-Al-Wadud, M., 2016. Optimal power allocation scheme for plug-in hybrid electric
vehicles using swarm intelligence techniques. Cogent Eng. 3 (1), 1203083.
Verzijlbergh, Remco, Lukszo, Z., Ilic, Marija D., 2012. Comparing different EV charging strategies in liberalized power systems. In: 2012 9th International Conference
on the European Energy Market (EEM). IEEE, pp. 1–8.
Wang, Qinglong, Liu, Xue, Du, Jian, Kong, Fanxin, 2016. Smart charging for electric vehicles: a survey from the algorithmic perspective. IEEE Commun. Surv. Tutorials
18 (2), 1500–1517.
Wang, Xudong, Pi, Yibo, Tang, Aimin, 2017. Scheduling of electric vehicle charging via multi-server fair queueing. IEEE Trans. Parallel Distrib. Syst.
Wei, Wei, Liu, Feng, Mei, Shengwei, 2016. Charging Strategies of EV Aggregator Under Renewable Generation and Congestion: A Normalized Nash Equilibrium
Approach.
Wen, C.-K., Chen, J.-C., Teng, J.-H., Ting, P., 2012. Decentralized plugin electric vehicle charging selection algorithm in power systems. IEEE Trans. Smart Grid 3 (4),
1779–1789.
Wu, Fei, Sioshansi, Ramteen, 2017. A two-stage stochastic optimization model for scheduling electric vehicle charging loads to relieve distribution-system constraints.
Transp. Res. Part B: Methodol. 102, 55–82.
Wu, Xiaolan, Cao, Binggang, Wen, Jianping, Bian, Yansheng, 2008. Particle swarm optimization for plug-in hybrid electric vehicle control strategy parameter. In: IEEE
Vehicle Power and Propulsion Conference, 2008., pp. 1–5.
Wu, Chenye, Mohsenian-Rad, Hamed, Huang, Jianwei, 2012a. Vehicle-to-aggregator interaction game. IEEE Trans. Smart Grid 3 (1), 434–442.
Wu, D., Aliprantis, D.C., Ying, L., 2012b. Load scheduling and dispatch for aggregators of plug-in electric vehicles. IEEE Trans. Smart Grid 3 (1), 368–376.
Wu, Tony Hao, Pang, Grantham K.H., Choy, K.L., Lam, H.Y., 2015. An optimization model for a battery swapping station in Hong Kong. In: Transportation
Electrification Conference and Expo (ITEC). IEEE, pp. 1–6.
Yan, Li, Wang, Jin-kuan, Han, Peng, Han, Ying-hua, 2013. Modeling and analysis on coordinated scheduling of E-bus recharging station participated wind-power
generation. In: Control Conference (CCC), 2013 32nd Chinese, pp. 8592–8596.
Yang, L., Zhang, J., Poor, H.V., 2014. Risk-aware day-ahead scheduling and real-time dispatch for electric vehicle charging. IEEE Trans. Smart Grid 5 (2), 693–702.
Yang, Hongming, Deng, Youjun, Qiu, Jing, Li, Ming, Lai, Mingyong, Dong, Zhao Yang, 2017. Electric vehicle route selection and charging navigation strategy based on
crowd sensing. IEEE Trans. Industr. Inf.
Yuqing, Yang, Jiuchun, Jiang, Yan, Bao, Weige, Zhang, Mei, Huang, Su Su. Dynamic coordinated charging strategy and positive effects in regional power system. In:
11th IEEE International Conference on Control & Automation (ICCA), 2014, pp. 1316–1321.
Zakariazadeh, Alireza, Jadid, Shahram, Siano, Pierluigi, 2014. Multi-objective scheduling of electric vehicles in smart distribution system. Energy Convers. Manage.
79, 43–53.
Zhan, Kaiqiao, Song, Yonghua, Hu, Zechun, Xu, Zhiwei, Jia, Long, 2012. Coordination of electric vehicle charging to minimize active power losses. In: Dianji,
Zhongguo, Xuebao, Gongcheng, (Eds.), Proceedings of the Chinese Society of Electrical Engineering, vol. 32, no. 31. Chinese Society for Electrical Engineering, pp.
11–18, 2012.
Zhan, Kaiqiao, Hu, Zechun, Song, Yonghua, Lu, Ning, Xu, Zhiwei, Jia, Long, 2015. A probability transition matrix based decentralized electric vehicle charging method
for load valley filling. Electr. Power Syst. Res. 125, 1–7.
Zhang, Peng, Qian, Kejun, Zhou, Chengke, Stewart, Brian G., Hepburn, Donald M., 2012. A methodology for optimization of power systems demand due to electric

416
M. Amjad et al. Transportation Research Part D 62 (2018) 386–417

vehicle charging load. IEEE Trans. Power Syst. 27 (3), 1628–1636.

Zhang, Tian, Chen, Wei, Han, Zhu, Cao, Zhigang, 2014a. Charging scheduling of electric vehicles with local renewable energy under uncertain electric vehicle arrival
and grid power price. IEEE Trans. Veh. Technol. 63 (6), 2600–2612.
Zhang, Kangkang, Xu, Liangfei, Ouyang, Minggao, Wang, Hewu, Lu, Languang, Li, Jianqiu, Li, Zhe, 2014b. Optimal decentralized valley-filling charging strategy for
electric vehicles. Energy Convers. Manage. 78, 537–550.
Zhang, Zhen, Pang, Hongliang, Lee, Christopher H.T., Xu, Xiangkui, Wei, Xile, Wang, Jiang, 2017. Comparative analysis and optimization of dynamic charging coils for
roadway-powered electric vehicles. IEEE Trans. Magn. 53 (11), 1–6.
Zhao, Qiang, Wang, Jinkuan, Han, Yinghua, Qian, Xiaolong, 2015. Research on unit commitment considering battery swapping station and fluctuating wind power. In:
Control and Decision Conference (CCDC), 2015 27th Chinese, pp. 5022–5027.
Zhou, K., Cai, L., 2014. Randomized PHEV charging under distribution grid constraints. IEEE Trans. Smart Grid 5, 879–887.
Zhu, Yuan, Chen, Yaobin, Tian, Guangyu, Wu, Hao, Chen, Quanshi, 2004. A four-step method to design an energy management strategy for hybrid vehicles. In:
American Control Conference, 2004. Proceedings of the 2004, vol. 1, pp. 156–161.

417