Full length article
Locating electric vehicle charging stations with service capacity using the improved whale optimization algorithm

https://doi.org/10.1016/j.aei.2019.02.006Get rights and content

Highlights

  • This paper presents an improved whale optimization algorithm (IWOA). On the basis of the whale optimization algorithm (WOA), this paper improves the algorithm by introducing a Gaussian operator, and meanwhile, mixing the differential evolution algorithm and the idea of crowding factor in the rear-end behavior of artificial fish swarm algorithm. It can be seen in the testing result that the IWOA significantly improve precision and computing speed of WOA.

  • In the base of the traditional study about EVs charging station siting, this paper takes service capacity factors into consideration, building charging stations with proper service capacity. In this model charging stations can be established with proper service capacity according to the local requirements, to reduce the variable costs and operating costs of setting charging stations in overall network.

  • This model considers secondary transport, namely power plant to charging station, and charging station to user's power transportation; the power transmission loss between the power plant and the charging station is taken into account. The applicability of this model has great significance for selecting the station location of EV charging station, making the charging station's location to be more reasonable and conform to the reality.

Abstract

This study proposes an Improved Whale Optimization Algorithm (IWOA) that, on the basis of Whale Optimization Algorithm (WOA) designed by Mirjalili and Lewis (2016), introduces Gaussian mutation operator, differential evolution operator, and crowding degree factor to the algorithm framework. Test results with nine classic examples show that IWOA significantly improves WOA’s precision and computing speed. We also model the locating problem of Electric Vehicle (EV) charging stations with service risk constraints and apply IWOA to solve it. This paper introduces service risk factors, which include the risk of service capacity and user anxiety, establishing the EV charging station site selection model considering service risk. Computational results based on a large-scale problem instance suggest that both the model and the algorithm are effective to apply in practical locating planning projects and help reduce social costs.

Introduction

With the booming of the automobile industry, many problems emerge as well, the most serious of which include air pollution and ground traffic congestion. Smart growth and electric vehicles are potential solutions to the negative impacts on air and health caused by worldwide urbanization [1]. The development of new energy EVs not only alleviates traffic pressure but also makes a great contribution to the improvement of atmospheric environment and energy conservation as well as emission reduction. As the basic supporting facilities for EV operation, charging stations have gradually stepped into a large-scale, networked planning and construction process along with the popularization of EV. For planners and managers, how to achieve the rational planning of charging station location, how to ensure the effective operation of the EV charging station network, how to realize smooth traffic, provide convenience for users’ travel and improve efficiency of the charging stations, are the problems to be addressed. The site selection of EV charging station must take factors such as urban planning, power grid planning, user demand and EV operation mode into consideration, which is a complex optimization problem. Moreover, this process will further promote the development of urban logistics and ultimately promote the rapid growth of metropolitan economy [2]. Therefore, the reasonable selection of construction positions for charging stations is of great significance.

Based on the traditional study concerning EV charging station siting, this paper considers service capacity factors, including the risk of service capacity and user anxiety. Service capacity [3] refers to the total charging quantity of the charging station that can meet the EV’ battery within a day. Because the charging demand in different areas varies, misconfiguring the service capacity introduces risks such as unmet demands and high costs. Meanwhile, building charging stations with proper service capacity will reduce the total cost of charging station construction. User anxiety refers to the user's fear of running out of power on the way to the charging station. To meet users’ needs and realize comprehensive cost minimization, the optimal site planning model is established. The swarm intelligence optimization algorithm is mostly used to solve the location problem, among which the whale optimization algorithm (WOA) is a new meta-heuristic optimization algorithm derived from Mirjalili and Lewis [4] in 2016 to solve optimization problems. WOA has few parameters, simple principles, strong global convergence and obvious advantages in convergence speed and precision. As a new heuristic optimization algorithm, WOA has wide application prospects. In this paper, three parts of improvement have been proposed, including introducing a Gaussian operator, combining the idea of the differential evolution algorithm and the crowded degree factor in the rear-end behavior of artificial fish swarm algorithm, which has greatly improved its optimization ability of WOA. The improved whale algorithm for model solving is used to study the best deployment scheme of public EV charging stations. Finally, the simulation results show that the model is effective and feasible.

Section snippets

Literature review

At present, studies on the theory of the location planning of EV charging facilities are still in the exploration stage. Research in related fields focuses on the development and existing problems of EVs, the design and site selection of EV charging facilities. Georg Brandstätter, Michael Kahr and Markus Leitner [5] introduced and studied the locating strategy optimization by using an EV sharing system. By taking two stages of the stochastic optimization method, they developed and tested a

Whale Optimization Algorithm (WOA)

The Whales optimization algorithm was a new swarm intelligence optimization algorithm put forward by Mirjalili and Lewis in 2016 [4]. By observing whale's social behavior, they found that whales have a special hunting method, which is called the bubble-net hunting method. The whale optimization algorithm, which simulates whale hunting, is divided into three stages: circling hunting, bubble-net attacking, and hunting for prey.

The locating model of EV charging station considering service risk

In order to meet the EV charging demand in city, according to the layout characteristics of urban public facilities, combined with the principles of convenient, economical, safe and feasible construction, alternatives for the electric vehicle charging station site selection are chosen. At last the optimal solutions are chosen from these alternatives to meet the demand. Due to the limitations of the project budget, how to minimize the total cost of the construction site in the case of satisfying

The example analysis

Suppose that a planning area is going to establish a network of EV charging stations considering service risk.

Conclusions

On the background of the EVs development, this paper studies the location of charging stations for EVs, and builds a charging station location model considering service risk. In this model charging stations can be established with proper service capacity in compliance with the local requirements, to reduce the variable costs and operating costs of setting charging stations in the overall network, At the same time, it can reasonably avoid users' anxiety because they are worried about the power

Acknowledgment

Thanks to the anonymous reviewers for their valuable comments.

Thanks to the strong support provided by the project of Beijing Philosophy and Social Science (NO.17GLB013) and the project of Research Base of Capital Distribution Industry (No. JD-YB-2018-006).

References (40)

  • L.D.S. Coelho et al.

    Differential evolution optimization combined with chaotic sequences for image contrast enhancement

    Chaos Solitons Fractals

    (2009)
  • Y.W. Wang et al.

    Locating road-vehicle refueling stations

    Transport. Res. Part E Logistics Transport. Rev.

    (2009)
  • Z. Huang et al.

    Problem of locating electric vehicle refueling stations with service capacity

    Ind. Eng. Manage.

    (2015)
  • D. Efthymiou et al.

    Electric vehicles charging infrastructure location: a genetic algorithm approach

    Eur. Transport Res. Rev.

    (2017)
  • A. Rajabi-Ghahnavieh et al.

    Optimal zonal fast-charging station placement considering urban traffic circulation

    IEEE Trans. Veh. Technol.

    (2017)
  • M. Hosseini et al.

    A heuristic algorithm for optimal location of flow-refueling capacitated stations

    Int. Trans. Operat. Res.

    (2017)
  • M.M. Islam et al.

    Optimal siting and sizing of rapid charging station for electric vehicles considering Bangi city road network in Malaysia

    Turkish J. Elec. Eng. Comp. Sci.

    (2016)
  • M. Hosseini et al.

    Selecting optimal location for electric recharging stations with queue

    Ksce J. Civil Eng.

    (2015)
  • J.E. Bell et al.

    An ant colony optimization approach to the vehicle routing problem

    Adv. Eng. Inf.

    (2003)
  • I. Aljarah et al.

    Optimizing connection weights in neural networks using the whale optimization algorithm

    Soft. Comput.

    (2016)
  • Cited by (95)

    View all citing articles on Scopus
    View full text