摘要
针对现有基于工况法的纯电动汽车续航里程预测算法精度难以提高的问题,提出一种基于数字地图的实时车流信息的纯电动车剩余续航里程在线预测算法。该算法使用EM机器学习算法对历史数据进行聚类分析,并建立一个支持向量机(SVM)模型。此模型使用在线数字地图(百度地图)的实时车流信息来预测车辆在给定的各路段的未来行驶工况,包括能量消耗与时间消耗。进而根据电池内阻模型估算电池SOC的变化量与剩余续航里程。最后在互联网分布式实车在环平台进行了一个约20 km的实车上路实验。实验结果证明了本方法的在线续驶里程预测有很高的准确度。
Aiming at the problem that the accuracy of the existing pure electric vehicle mileage prediction algorithm is difficult to improve,an online prediction algorithm for the remaining cruising range of pure electric vehicles based on real-time traffic flow information is proposed.The algorithm uses the EM machine learning algorithm to cluster historical data and establish a support vector machine(SVM)model.It uses the real-time traffic flow information of the online digital map(Baidu map)to predict the future driving conditions on each given road section,including energy and time consumption.Then,the battery SOC variation and remaining cruising range are estimated.Finally,a 20 km real-vehicle on-road experiment was conducted on the Internet-distributed vehicle-in-the-loop platform.Experimental results prove that this online mileage prediction method has high accuracy.
作者
冉勇川
曾军
张毅
RAN Yongchuan;ZENG Jun;ZHANG Yi(School of Vehicle Engineering,Chongqing University of Technology,Chongqing 400054,China)
出处
《机械科学与技术》
北大核心
2025年第8期1442-1450,共9页
Mechanical Science and Technology for Aerospace Engineering
基金
重庆市教委青年基金项目(KJQN202001105)。
作者简介
冉勇川,本科生,1918127143@qq.com;通信作者:张毅,讲师,博士,zagyi81@cqut.edu.cn。
