基于ISODATA的实时动态电动汽车运行工况与耗电特性研究  

RESEARCH ON OPERATING CONDITIONS AND POWER CONSUMPTION CHARACTERISTICS OF REAL-TIME DYNAMIC ELECTRIC VEHICLES BASED ON ISODATA

作　　者：李广玮 吴鸣[2] 王昕扬 徐毅 Li Guangwei;Wu Ming;Wang Xinyang;Xu Yi(Shanghai University of Electric Power,Shanghai 200082,China;China Electric Power Research Institute,Beijing 100192,China)

机构地区：[1]上海电力大学,上海200082 [2]中国电力科学研究院,北京100192

出　　处：《计算机应用与软件》2021年第11期344-349,共6页Computer Applications and Software

摘　　要：针对电动汽车行驶工况进行研究是确定电动汽车能耗、电动汽车新型技术和评估的核心方法。采集电动汽车在实际道路行驶的数据,使用主成分分析算法(Principal Component Analysis, PCA)、迭代自组织数据分析算法(Iterative Self Organizing Data Analysis Techniques Algorithm, ISODATA)以及运动学片段分析法对实测的数据进行降维和聚类,利用Silhouette函数验证聚类结果的合理性。根据聚类中心的大小,筛选提取的运动学片段,构建电动汽车实际行驶的代表性工况,通过测试数据进行了差异性检验。在建立代表性工况的基础上,提出电动汽车能耗特性和电量实时估算方法。Research on electric vehicle driving conditions is the core method for determining electric vehicle energy consumption, new technologies and evaluation of electric vehicles. This article collected data of electric vehicles driving on actual roads, and used principal component analysis(PCA), iterative self-organizing data analysis techniques algorithm(ISODATA), and kinematic segment analysis to realize the dimension reduction and clustering of measured data. The Silhouette function was used to verify the rationality of the clustering results. According to the size of the clustering center, the extracted kinematic fragments were screened, and the representative operating conditions of electric vehicles were constructed. The difference test was carried out through test data. Based on the establishment of representative working conditions, a real-time estimation method for energy consumption characteristics and power consumption of electric vehicles is proposed.

关 键 词：电动汽车 行驶工况 耗电特性 迭代自组织数据分析算法 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]