长时间尺度下计及里程焦虑心理效应的电动汽车充放电调度策略  

Electric Vehicle Charging and Discharging Scheduling Strategy Considering Psychological Effect of Mileage Anxiety on a Long-time Scale

作　　者：侯慧[1] 王逸凡 吴细秀[1] 陈跃 黄亮[1] 张锐明 HOU Hui;WANG Yifan;WU Xixiu;CHEN Yue;HUANG Liang;ZHANG Ruiming(School of Automation,Wuhan University of Technology,Wuhan 430070,China;Guangdong Guangshun Renewable Energy Technology Co.,Ltd.,Foshan 528000,China)

机构地区：[1]武汉理工大学自动化学院,武汉430070 [2]广东广顺新能源动力科技有限公司,佛山528000

出　　处：《高电压技术》2023年第1期85-93,共9页High Voltage Engineering

基　　金：国家重点研发计划(2018YFB0105700);国家自然科学基金(52177110);深圳市科技计划(JCYJ20210324131409026)。

摘　　要：电动汽车在需求侧能够发挥出优秀的灵活响应潜力。以往研究多侧重电动汽车的短时间尺度调度,但短时间尺度调度未从宏观时间尺度优化而易陷入短时局部最优。为此,提出了一种长时间尺度下计及里程焦虑心理效应的电动汽车充放电调度策略。首先,构建了长时间尺度电动汽车最优充放电调度策略框架;其次,考虑了电动汽车充放电操作的电池损耗成本以及电动汽车用户出行里程焦虑心理效应,以完善对电动汽车用户效益的量化;在此基础上,建立了考虑电动汽车调度成本及里程焦虑心理效应的长时间尺度电动汽车日前-实时双层多目标调度模型;最后,基于滚动时域优化方法对实时优化问题进行处理,利用非支配排序遗传算法对多目标问题进行求解。算例表明:所提策略可使电动汽车用户的里程焦虑心理效应在调度周期内长期维持低于0.25;维持较高水平(如0.75~0.9)的荷电状态可降低调度成本及里程焦虑,提升用户满意度,进而提升其接受调度的积极性与参与度。Electric vehicles(EVs) can exert excellent demand response potential on the demand side. However, previous studies have mostly focused on the short-time scheduling of EVs, which fails to optimize from a macro-time scale and thus easily falls into short-time local optimization. To solve the problem, this paper proposed a charging and discharging scheduling strategy for EVs considering the psychological effect of mileage anxiety on a long-time scale. Firstly, the framework of the optimal EV charging and discharging scheduling strategy of long-time scale was constructed. Secondly,the cost of battery loss of vehicle-to-grid operation and the psychological effect of mileage anxiety of EV users were considered to improve the quantification of the EV users’ benefits. On this basis, a long-time-scale day-ahead-real-time two-layer multi-objective scheduling model for EVs was established, in which the scheduling cost and the psychological effect were taken into consideration. Finally, the rolling horizon optimization method was used to solve the real-time optimization problem and the non-dominated sorting genetic algorithm II was used to solve the multi-objective problem.Simulation cases show that the proposed strategy can maintain the psychological effect of mileage anxiety of EV users within 0.25 for a long time in the scheduling cycle. Maintaining a high-level state of charge(for example within 0.75~0.9)can reduce the scheduling cost and the mileage anxiety, improve users’ satisfaction, and thus enhance their enthusiasm and participation in scheduling.

关 键 词：长时间尺度 心理效应 电动汽车 充放电 滚动时域 

分 类 号：U491.8[交通运输工程—交通运输规划与管理] TM73[交通运输工程—道路与铁道工程]