A bi-objective model for location planning of electric vehicle charging stations with GPS trajectory data. (February 2019)

• Record Type: Journal Article
• Authors: Bai, Xue; Chin, Kwai-Sang; Zhou, Zhili
• Is Part Of: Computers & industrial engineering. Volume 128(2019)
• Page Start: 591
• Page End: 604
• Publication Date: 2019-02
• Subjects: Electric vehicles -- Charging infrastructure location -- Bi-objective optimization -- Bi-objective Evolutionary algorithm -- GPS trajectory data; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2019.01.008
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: This paper addresses the problem of designing a charging station network in the city with low EV penetration rate. We present a cell-based model to decide locations, capacity options, and service types for EV charging stations. The hybrid evolutionary algorithm is developed and evaluated. A case study of Shenzhen, China is presented with GPS trajectory data from thousands of vehicles. Abstract: The construction of charging stations is a crucial factor in promoting electric vehicles (EV). It is necessary to construct EV charging stations in advance to encourage drivers to prefer EVs. This paper addresses the EV charging stations location problem in a city with low EV penetration rate. We divide the city into a grid with several same cells. The potential charging demand of each cell is estimated with the use of GPS trajectory data from thousands of traveling vehicles in the network. We present a cell-based model to decide locations, capacity options, and service types for EV charging stations that can cover all potential charging demand. The problem is formulated as a bi-objective mixed-integer mathematical model, with one objective related to minimizing cost and the other related to maximizing service quality. To solve it, we propose a hybrid evolutionary algorithm that combines the non-dominated sorting genetic algorithm-II (NSGA-II) with linear programming and neighborhood search. We conduct computational experiments on randomly generated instances to evaluate the performance of the proposed hybrid NSGA-II. Finally, we present a case study designing an EV charging station network for Shenzhen, China with real GPS trajectory data. We also offer some management insights of EV charging stations construction based on sensitivity analysis.