基于单亲遗传算法混合动态规划的电动汽车充电调度优化策略  

An optimal charging schedule strategy of electric vehicles based on partheno-genetic algorithm and dynamic programming

作　　者：陆坚毅[1] 杨超[1] 肖来元[2] 郑锐[1] 

机构地区：[1]华中科技大学管理学院,湖北武汉430074 [2]华中科技大学软件学院,湖北武汉430074

出　　处：《计算机工程与科学》2015年第5期967-973,共7页Computer Engineering & Science

基　　金：国家自然科学基金重大项目资助(71320107001);中央高校基本科研业务专项资金资助项目(HUST:2013QN101;2013ZZGH028)

摘　　要：充电调度是电动汽车运营的一个重要内容,合理有效的充电策略在帮助运营商降低成本的同时还能减轻电网高峰时段的供电负担。从充电站运营商的角度出发,在实时电价和每个充电任务时间必须连续的假设下,建立了一个电动汽车充电成本最小模型,给出一个单亲遗传算法混合动态规划的两阶段常规充电调度算法。与电桩一旦闲置即刻分配车辆进行充电的策略以及传统单亲遗传算法相比较,该充电调度策略在电桩负载均衡的情况下有效降低了电费成本,说明了算法的有效性。此外,实验结果表现出了充电任务在多数相同时段聚集从而避开高电价时段的特征,说明充电策略对减轻高峰时段的电网压力也有一定帮助。Battery charging scheduling is one of the most important aspects in electric vehicle opera- tions management. An efficient charging scheme can not only help operator reduce the charging cost but also relieve the stress of the power system during peak hours. Based on the assumptions of real time elec- tricity price and no-interruption charging jobs,we propose a minimum charging cost strategy on the basis of partheno-genetic algorithm and dynamic programming. To test the performance, we make a compari- son between our algorithm and a designed strategy called “first come first charge” and the traditional ge- netic algorithm. We test the three charging strategies with the same examples and the simulation results indicate that the proposed method is effective in cost saving while insuring the loading balance of the e- lectric system. In addition,the Gantt chart of vehicle assignment also shows it can effectively relieve the stress of the grid.

关 键 词：电动汽车 充电调度 单亲遗传算法 动态规划 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]