报告题目 | Data-driven energy-efficient reliable path finding algorithm for electric vehicles |
报告人（单位） | 吴婷教授（南京大学） |
主持人（单位） | 李泳臻（太阳集团tyc539） |
会议时间 | 2024年09月18日 20:30 |
会议地点 | 腾讯会议号：163-340-929 会议设有密码，请参见线下展板或与主持人联系。 提醒：请参会者以真名进入，否则可能会被移出会议！ |
报告人简介 | |
吴婷，南京大学教授，博导，数学学院副院长，江苏高校青蓝工程优秀青年骨干教师，江苏“六大人才高峰”高层次人才。担任江苏运筹学会常务理事，江苏大数据联盟常务理事。学科领域为运筹学，长期从事大数据分析、最优化理论及其应用等方面的教学、科研与实践工作。先后主持了国家自然科学基金、中国博士后科学基金特别项目、江苏省自然科学基金，江苏省软科学项目等十余项国家级或省部级科研项目。在国际权威期刊上发表学术论文近30篇，并出版专著1部，获批专利1项。 | |
报告内容提要 | |
In this paper, we develop a novel reliable path finding algorithm for a stochastic road network with uncertainty in travel times while both electric vehicle energy and efficiency are simultaneously taken into account. We first propose a bi-objective optimization model to maximize (1) the on-time arrival reliability and (2) energy-efficiency for battery electric vehicles (BEVs) in a path finding problem. The former objective requires finding the reliable shortest path (RSP), which is the path with the minimal effective travel time measured by the sum of the mean travel time and a travel time safety margin for any given origin-destination (OD) pair. Then, we refer to energy-efficiency as the minimum of the electric energy consumption. We discuss the non-additive property of the RSP problem since we also consider the link travel time correlations, whereas the latter objective satisfies the additive criterion. To this end, we illustrate the existence of non-dominated solutions that satisfy both of the two objectives. Furthermore, it is shown that the intersection of two candidate sets – one for the RSPs and the other for paths with minimal energy-consumption - actually contains the optimal solution for the bi-objective optimization problem. The upper and lower bounds of the effective travel time are mathematically deduced and can be used to generate the candidate path set of this bi-objective problem via the K-shortest algorithm. Our proposed algorithm overcomes the infeasibility of traditional path finding algorithms (e.g., the Dijkstra algorithm) for RSPs. Moreover, using two numerical examples, we verify the effectiveness and efficiency of the proposed algorithm. We numerically demonstrate promising potential applications of the proposed algorithm in real-life road traffic networks.
| |
