基于主成分和BP神经网络的行驶工况辨别研究  

Study on Identification of Driving Cycle Based on Principal Component and Back Propagation Network

作　　者：王楠楠 徐小东[1] 李建松[2] WANG Nan-nan;XU Xiao-dong;LI Jian-song(Anhui Communications Vocational&Technical College,Hefei 230051,China;Xuzhou College of Industrial Technology,Xuzhou 221140,China)

机构地区：[1]安徽交通职业技术学院,安徽合肥230051 [2]徐州工业职业技术学院,江苏徐州221140

出　　处：《长春师范大学学报》2020年第6期35-41,共7页Journal of Changchun Normal University

基　　金：安徽高校自然科学研究项目资助“基于行驶工况辨识的P2型PHEV能量管理策略研究”(KJ2018A0782);安徽高校自然科学研究项目资助“汽车自动变速器实验台架研究”(KJ2018A0786);2018-2020年度中国交通教育研究会教育科学研究课题“基于常态化教学诊断与改进的中职汽修专业教学研究与实践”(交教研1802-244)。

摘　　要：针对当前行驶工况识别算法中由于数据预处理不足导致的识别精度较低等问题,将多元统计理论中的主成分分析算法引入到数据的预处理过程中,建立基于主成分和BP神经网络的行驶工况辨别模型。在代表三种行驶工况类型的多条道路上进行实车试验,采集大量行驶工况数据作为实证研究的数据源,最终通过均方误差函数和混淆矩阵验证了识别模型的性能和精度。研究结果表明,通过主成分分析和BP神经网络相结合的方法可有效提高识别精度,识别模型对混合动力汽车调整能量控制策略,实现实时最优控制具有较大的意义。In order to solve the problem of low recognition accuracy caused by insufficient data pre-processing in current recognition algorithm of driving cycle,this paper introduces principal component analysis(PCA)algorithm of multivariate statistical theory into the data preprocessing process.A driving condition identification model based on principal component analysis and BP neural network is established.A large number of road test data were collected on several roads representing three driving modes as the data source of the empirical study.Finally,the performance and accuracy of the recognition model were verified by the mean square error function and the confusion matrix.The results show that a high accuracy was got by the combination of principal component analysis and BP neural network.There is great significance for hybrid vehicles to adjust energy control strategies and achieve real-time optimal control.

关 键 词：行驶工况 主成分分析 BP神经网络 

分 类 号：U469.72[机械工程—车辆工程]