Optimization design of electric vehicle charging stations based on the forecasting data with service balance consideration

摘要

As fast growth of Electric Vehicle(EV) and lag of charging station construction in China, this paper introduces an optimization model of charging stations design, where the parameters of EV number are forecasted by data analysis with Nonlinear Auto-Regressive neural network. Firstly, the procedure of EV forecasting is introduced. Secondly, the mathematical model is proposed to optimize the layout of EV stations and the number of equipment, with consideration of balance service that includes idle rate and waiting time limits. In the model, each station is simulated as a queuing system. Then, as the model is a NP-hard problem, a hybrid heuristic algorithm that combines Genetic Algorithm and Binary Particle Swarm Optimization is proposed. Through a large scale of computational examples, the algorithm is proved to be more efficient than Genetic Algorithm. Finally, the performance of the method is illustrated by a case study. (C) 2018 Elsevier B.V. All rights reserved.