电动汽车城市道路行驶工况特征参数智能混合搜索算法研究  

Research on intelligent hybrid search algorithm for urban road driving cycle characteristic parameters of electric vehicles

作　　者：吴若园 罗文广 覃永新 蓝红莉 庞娜 WU Ruoyuan;LUO Wenguang;QIN Yongxin;LAN Hongli;PANG Na(School of Electrical and Information Engineering,Guangxi University of Science and Technology,Liuzhou 545006,China;Guangxi Key Laboratory of Auto Parts and Vehicle Technology,Liuzhou 545006,China)

机构地区：[1]广西科技大学电气与信息工程学院,广西柳州545006 [2]广西汽车零部件与整车技术重点实验室,广西柳州545006

出　　处：《重庆理工大学学报（自然科学）》2022年第8期36-44,共9页Journal of Chongqing University of Technology：Natural Science

基　　金：广西自然科学基金重点项目(2020GXNSFDA238011);广西自动检测技术与仪器重点实验室开放基金项目(YQ21203);广西汽车零部件与整车技术重点实验室基金项目(2020GKLACVTZZ02)。

摘　　要：为了更好地解决行驶工况特征参数选择的问题,提出了一种将粒子群算法与禁忌搜索算法相结合的智能混合搜索算法。在对实车采集的数据进行预处理的基础上划分运动学片段,并按照电动汽车城市道路行驶的四种工况对片段进行分类,作为智能混合搜索算法中分类器的训练数据集和测试数据集。使用粒子群算法对行驶工况特征参数的分段参数部分的参数边界进行搜索和优化,由此计算行驶工况特征参数全集,使用禁忌搜索算法进行全集搜索,选择行驶工况特征参数最优子集。结果表明:提出的算法选择11个特征参数组成的最优子集使随机森林分类器的识别准确度达到87.74%的较高水准,且能在较短计算时间内选择特征参数最优子集。An intelligent hybrid search algorithm combined with the particle swarm algorithm and the taboo search algorithm is proposed to solve the problem of selecting driving cycle characteristic parameters much better. On the basis of preprocessing the data collected by real vehicles, the kinematic segments are divided, and the segments are classified according to the four driving conditions of electric vehicles on urban roads, which are used as the training data set and test data set of the classifier in the intelligent hybrid search algorithm. The particle swarm optimization algorithm is used to search and optimize the boundaries between distribution measures of driving cycle parameters, so as to calculate the universal set of driving cycle parameters. The universal set is searched by taboo search algorithm, and the optimal subset of driving cycle parameters is selected. The results show that the proposed algorithm selects out the optimal subset with 11 characteristic parameters, which makes the recognition accuracy of the random forest classifier reach a high level of 87.74% in a short calculation time.

关 键 词：最优子集 智能混合搜索算法 行驶工况特征参数 粒子群算法 禁忌搜索算法 电动汽车 城市道路 

分 类 号：U469.72[机械工程—车辆工程]