摘要
城市区域共享型电动汽车(ECSS)优化调度需要解决越发突出的用户用车需求与网点车辆供需不平衡、网点车辆充电与电网运行状态不匹配等问题。该文提出面向共享电动汽车的用户助推方案与充电优化的协同调度方法。首先,设计融合了认知和动机助推策略框架的面向强、弱助推场景的用户助推方案,并采用披露-模糊综合评价方法(r-FCEM)评估助推方案的用户响应度,通过助推引导用户还车至理想网点以解决车辆平衡调度问题;然后,基于计及经济性和低碳性的共享电动汽车充电优化模型,设计助推和充电优化协同调度下的ECSS环境,与深度Q网络(DQN)智能体交互,通过模拟用户-运营商-电网三者交互,对车辆的助推方案和充电优化策略进行协同求解,以解决充电匹配问题;最后,通过调研和算例,验证了所提方法能在满足用户选择自由权的基础上以非强制策略提高用户响应度,提升了运营商经济效益,降低了电网负荷波动。
Electric car-sharing(ECS),as a component of the sharing economy,is of great significance in alleviating urban traffic congestion and reducing carbon emissions.Electric car-sharing system(ECSS)involves multiple entities such as users,operators and power grids.At present,one-way network operation mode is mostly adopted.Users can pick up and return vehicles at any network specified by the operator,and the operator arranges for vehicles in the network to connect to the power grid for charging.The optimal scheduling of urban electric car-sharing system is needed to solve the increasingly prominent problems such as imbalance between user demand and station cars supply,and mismatch between cars charging and grid operation status.Current strategies for vehicle scheduling are high-cost and coercive,while charging scheduling only ensures vehicle availability,lacking consideration of the impact of vehicle charging on the grid.Addressing these issues,the application of low-cost,non-coercive nudging methods from behavioral economics in the field of ECSS was explored and a coordinated user nudging and charging optimization scheduling method for urban shared electric vehicles was proposed.Firstly,at the level of vehicle scheduling with supply and demand balance,nudging was used to guide user dispatch.Based on actual surveys,the main factors influencing users'choice of return points were identified,and nudging schemes for strong and weak scenarios were designed based on a framework of motivational and cognitive nudges.The revealed fuzzy comprehensive evaluation method(r-FCEM)was used to evaluate the user responsiveness to the nudging schemes,determining the probability of users participating in vehicle dispatch,thereby relocating vehicles from surplus supply points to stations with high demand,and improving operators'rental service income.And then we tested the feasibility of the nudging scheme and found that the design of the nudging scheme for users'choice of return stations can effectively improve user responsiveness and has a certain degree of feasibility.Secondly,for the charging scheduling problem,nudge guided users to return vehicles to low-cost,low-carbon stations,and charging optimization model considering economic and low-carbon factors was designed.Based on deep Q network(DQN),an ECSS operating environment was constructed to simulate the interactions among users,operators,and the grid.After training process,coordinated solutions for nudging and charging optimization were obtained.This resulted in a dispatch plan for vehicle scheduling and a charging schedule for charging optimization.The research first examined the number of vehicles and the travel and arrival volumes at typical stations under nudged and non-nudged scenarios,demonstrating the impact of nudging on supply-demand imbalance and charging optimization issues.It was found that user nudging can alleviate phenomena of under-supply and surplus,guiding vehicles to low-cost,low-carbon stations.Then,four scenarios were set up,revealing that single vehicle scheduling and charging scheduling alone offer limited improvement to the economic benefits of ECSS.It is necessary to solve nudging and charging scheduling in a coordinated manner to enhance user responsiveness through non-coercive strategies,reduce grid load fluctuations,and comprehensively improve the economic efficiency of operators while addressing vehicle scheduling and charging optimization problems.Future work on nudging will expand the scope and number of questionnaire surveys to further validate the feasibility and effectiveness of practical applications.Algorithmically,future research will focus on refined modeling for large-scale ECSS operations and seek better algorithms to adapt to large-scale scenarios.
作者
陈中
万玲玲
张梓麒
Chen Zhong;Wan Lingling;Zhang Ziqi(School of Electrical Engineering,Southeast University,Nanjing 210096 China)
出处
《电工技术学报》
北大核心
2025年第11期3572-3590,共19页
Transactions of China Electrotechnical Society
基金
国家自然科学基金资助项目(52077035)。
关键词
共享电动汽车系统
车辆供需平衡
充电状态匹配
助推
充电控制
Electric car-sharing system(ECSS)
supply and demand balance of car
match of charging and grid status
nudge
charging control
