自动变速操纵系统稳态过程故障检测和诊断技术研究  

Research on Fault Detection and Diagnosis of Automatic Transmission Control System under Steady State Condition

作　　者：彭建鑫[1] 刘海鸥[1] 王滨[1] 陈慧岩[1] 

机构地区：[1]北京理工大学机械与车辆学院,北京100081

出　　处：《兵工学报》2013年第11期1352-1358,共7页Acta Armamentarii

基　　金：国家高技术研究发展计划项目(2011AAllA252)

摘　　要：采用多向主元分析(MPCA)算法实现了稳态工况下自动变速操纵系统(ASCS)的故障检测和诊断功能。针对稳态工况下ASCS故障诊断的问题,根据ASCS控制周期特点研究稳态工况下ASCS状态变量特性以及变量构成成分,分析了MPCA算法的可行性;运用稳态工况下无故障历史数据建立ASCS的MPCA模型,并采用综合监控指标OIndex进行过程故障检测。当故障发生时利用因子分析理论,建立了综合监控指标、得分向量、系统状态变量之间的映射关系,实现故障诊断功能;采用实车试验和仿真验证结合的方式证明了MPCA算法在ASCS稳态工况下故障检测和诊断的有效性和实时性。Fault detection and diagnosis of automatic transmission control system （ASCS） is realized by multi-way principal component analysis （MPCA）. According to fault diagnosis of ASCS under steady state condition, firstly, the state variable characteristics under steady state condition are analyzed and the feasibility of MPCA algorithm is researched by taking ASCS＇ s control cycle characteristics as the basis. Secondly, ASCS＇ s multi-way principal component models are established by faultless historical data. And the comprehensive monitoring indicator, OIndex, is used for process fault detection. When a fault occurs, a fault isolation is achieved by establishing a mapping relationship among the comprehensive monitoring indicator, score vectors and state variable characteristics. At last, the vehicle test and simulation test are used to prove the effectiveness and real-time of MPCA algorithm under ASCS＇ s steady state condition.

关 键 词：交通运输安全工程 自动变速操纵系统 多向主元分析 故障诊断 机械式自动变速器 

分 类 号：U463.2[机械工程—车辆工程]