电动汽车充电站规划的多种群混合遗传算法

Electric Vehicle Charging Station Planning Based on Multiple-population Hybrid Genetic Algorithm

建立考虑充电站建设运营成本和充电者充电成本的电动汽车充电站综合成本最小模型。针对电动汽车充电站规划的特点,提出了一种新的多种群混合遗传算法(MPHGA)。该算法将标准遗传算法(SGA)与交替定位分配算法(ALA)结合,针对充电站规划的多目标性,采用多种群概念,建立多种群并进行协同进化搜索。基于地理信息系统(GIS),考虑地理信息对充电站选址的影响,通过某市充电站规划实例验证了该模型和方法的正确性和有效性。

The electric vehicle charging station＇s minimum comprehensive cost model is established in the paper, which considers charging station construction and operation cost and the cost of charging people. According to the characteristics of the electric vehicle charging station planning, a new kind of multiple-population hybrid genetic al- gorithm（ MPHGA ）is proposed .The algorithm combines the standard genetic algorithm （SGA） with alternative location and allocation algorithm （ALA）. According to the multi-objective of the charging station planning, the concept of muhigroup is used to do collaborative evolution search. Based on the geographic information system （GIS）, the geo- graphic information influence on the location of the charging station will be considered. The model and method are proved that they have great correctness and effectiveness by a charging station planning example of a city.