基于聚类算法的汽车行驶工况构建方法研究  

Research on construction of vehicle driving cyclebased on clustering algorithms

作　　者：张益敏 吴小兰[1] 张奕河[1] 吕泓钊 缪嘉荣 ZHANG Yimin;WU Xiaolan;ZHANG Yihe;LV Hongzhao;MIAO Jiarong(Fujian Electric Power Vocational and Technical College,Quanzhou,Fujian 362000,China)

机构地区：[1]福建电力职业技术学院,福建泉州362000

出　　处：《计算机应用文摘》2024年第8期82-84,88,共4页Chinese Journal of Computer Application

基　　金：福建电力职业技术学院2023年校级科研项目:聚类算法在数据分析中的应用——以汽车行驶工况曲线构建问题为例(2023KY014)。

摘　　要：构建符合地区经济和交通发展状况的汽车行驶工况具有重要意义。文章采用主成分分析和聚类算法构建适应地区实际情况的汽车行驶工况。首先,根据问题实际对原始数据进行清洗,降低采集误差对结果的影响,再进行运动学片段划分;其次,采用主成分分析法和K-means聚类算法对运动学片段进行分类;最后,采用最小偏差采样方法分别选取每类的代表性片段以合成最终工况。文章所构建的工况的各特征值与原始数据之间的相对误差均值在5%以内,能够有效反映汽车实际行驶状况。It is of great significance to construct driving conditions that are in line with the economic and transportation development of the region.The article uses principal component analysis and clustering algorithm to construct a vehicle driving cycle that adapts to the actual situation in the region.Firstly,based on the actual problem,clean the original data to reduce the impact of collection errors on the results,and then divide the kinematic segments.Secondly,principal component analysis and K-means clustering algorithm are used to classify kinematic segments.Finally,the minimum deviation sampling method is used to select representative fragments of each class separately to synthesize the final operating conditions.The average relative error between the characteristic values of the working conditions constructed in the article and the original data is within 5%,which can effectively reflect the actual driving conditions of the car.

关 键 词：汽车行驶工况 短行程法 主成分分析 K-MEANS聚类 

分 类 号：U467[机械工程—车辆工程]