基于变权Voronoi图和混合粒子群算法的电动汽车充电站规划

针对城市电动汽车充电站规划布局及服务范围划分的问题,提出一种基于变权Voronoi图和混合离散粒子群算法的优化算法。为解决离散粒子群算法在达到最优解时容易发生变异的问题,改进了离散粒子群算法中的概率映射函数,提高算法迭代中后期的全局搜索能力;引入加权Voronoi图生成过程中可随充电站的服务能力和最大服务半径约束动态调整的变权重系数,使充电站服务范围的划分可控且更为合理;利用最短路径法求得用户充电行驶过程中的交通路径距离来取代传统的欧式距离,提高算法的准确性。运用改进后的混合离散粒子群-变权Voronoi图算法求解算例模型,通过算例结果验证了所提算法用于电动汽车充电站规划的有效性。

Optimal Planning of Charging Stations for Electric Vehicle Based on Weight-Changed Voronoi Diagram and Hybrid Particle Swarm Optimization Algorithm

In allusion to solve the optimal planning problem of charging stations for electric vehicle and division of services range,a method based on weight-changed voronoi diagram and hybrid binary particle swarm optimization(WCVD-HBPSO)Algorithm was proposed.The binary particle swarm optimization algorithm was improved by modifying the function of probability map to improve the global searching capability when the number of iterations reaching to the latter stage;Introducing a weight-changed coefficient which could be dynamically adjusted in the process of generating the improved Voronoi diagram according to the constraints of charging station's service capability and the maximum service radius,then the process of dividing service range is controllable and more reasonable;Using the shortest path algorithm to improve the accuracy of the algorithm by calculating the distance along the traffic path to replace the Euclidean distance of EV users running to the nearest station.The proposed WCVD-HBPSO algorithm were tested by a given example,the results verified its feasibility and effectiveness.