基于精英强化遗传算法的电动汽车充电调度优化  

Electric Vehicle Charging Scheduling Optimization Based on Elite Enhanced Genetic Algorithm

作　　者：谢涛 谭飞 黄军付 王俊佳 袁超杨 XIE Tao;TAN Fei;HUANG Junfu;WANG Junjia;YUAN Chaoyang(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Yibin 644000,China;Artificial Intelligence Key Laboratory of Sichuan Province,Yibin 644000,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川宜宾644000 [2]人工智能四川省重点实验室,四川宜宾644000

出　　处：《四川轻化工大学学报(自然科学版)》2025年第1期86-93,共8页Journal of Sichuan University of Science & Engineering(Natural Science Edition)

基　　金：国家自然科学基金项目(61902268);四川省科技计划项目(21ZDYF4052,2020YFH0124,2021YFSY0060);四川轻化工大学创新创业训练项目(CX2023198,CX2023193,CX2023195)。

摘　　要：针对区域内电动汽车(EV)如何实现有效调度大规模充电作业问题,提出了一种基于精英强化遗传算法(EEGA)的电动汽车充电调度优化算法。首先引入自适应策略动态改变交叉、变异概率,解决传统算法易陷入局部最优的问题;然后将父代与子代种群通过精英保留策略形成新的种群,并对新种群的最优个体进行强化搜索,寻找更优解替换,提高算法收敛速度;最后通过制定充电调度策略,建立多目标车辆调度模型,并采用熵权TOPSIS法消除多目标的维度。仿真结果表明,与传统遗传算法调度相比,充电时间、充电成本、充电桩利用偏差率和电网负荷分别降低了约4.2%、2.3%、4.4%和6.8%,所提出的优化方法提供了接近最优的解决方案,有效调度大规模电动汽车充电作业。In order to solve the problem of how to efficiently schedule charging tasks for electric vehicle(EV)in a large scale,an optimization method for EV charging scheduling based on elite enhanced genetic algorithm(EEGA)has been proposed.Firstly,an adaptive strategy is introduced to dynamically change crossover and mutation probability,which is used to solve the problem that the traditional algorithm is prone to fall into local optimum.Secondly,a new population is formed by the parent and child populations through the elite retention strategy,and the optimal individuals of the new population are searched to find a better solution replacement,which is helpful to improve the algorithm convergence speed.Lastly,the multi-objective vehicle scheduling model is established by making charging scheduling strategy,and entropy-weighted TOPSIS method is used to eliminate the multi-objective dimension.The simulation results show that the charging time,charging cost,charging pile utilization deviation rate and grid load are reduced by about 4.2%,2.3%,4.4%and 6.8%,respectively,compared with the traditional genetic algorithm scheduling.The proposed optimization method provides a near-optimal solution to effectively schedule large-scale EV charging operations.

关 键 词：电动汽车 调度优化 精英强化 遗传算法 熵权TOPSIS法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] U491.8[自动化与计算机技术—控制科学与工程]