汽车主动悬架传感器故障检测与隔离研究  

Study on Fault Detection and Isolation of Vehicle Active Suspension Sensor's Fault

作　　者：杨柳青[1,2] 陈无畏[1] 张卫华[1] 汪洪波[1] 

机构地区：[1]合肥工业大学机械与汽车工程学院,合肥230009 [2]安徽交通职业技术学院汽车与机械工程系,合肥230051

出　　处：《控制工程》2014年第2期283-289,共7页Control Engineering of China

基　　金：国家自然科学基金资助项目(51075112;51178158);安徽省高校省级自然科学研究项目(KJ2012B050)

摘　　要：针对汽车主动悬架(Active Suspension System,ASS)传感器卡死、增益变化、恒偏差常见故障,提出一种故障检测与隔离方法(Fault Detection and Isolation,FDI)。建立主动悬架4自由度半车模型和传感器故障时故障悬架模型,在利用故障检测滤波器获得主动悬架输出残差的基础上,设计故障检测指标,计算故障检测指标实时值并于阈值比较,实现传感器故障定量化检测。利用Kalman滤波器组获得主动悬架状态估计信息,选取合适二级决策变量构造故障隔离决策函数,根据决策函数对故障响应敏感情况隔离故障传感器。Matlab/Simulink仿真实验结果与分析表明:故障检测指标实时值大于等于阈值0.5时,可定量化检测出主动悬架传感器故障;同时,相对其他二级决策变量出现明显大幅跳跃性波动的二级决策变量,对应传感器被隔离为故障传感器。该方法能有效实现汽车主动悬架传感器故障检测与隔离,优化悬架设计,提高车辆控制可靠性和使用安全性。A sensor fault detection and isolation method was put forward for ordinary faults as vehicle active suspension system dead- locking, gain variation and constant deviation. A four-degree-of-freedom half vehicle model with active suspension system was built, and so was a fault suspension model when the sensor failed. Sensor＇ s fault quantitative detection was realized by output residual error obtained from fault detecting filter, by designing the fault detection index and by calculating the real-time value of fault detection index and comparing it with the threshold value. Active suspension state estimation information was obtained from Kalman filter group, mean- time the fault isolation decision-making function was constructed by selecting the appropriate two-stage decision variables, thus to isolate faulty sensor according to the response sensitivity of decision-making function to the faults. The results of simulation experiment and a- nalysis under Matlab/simulink show that the sensor fault can be detected quantitatively when fault detection index real-time value is greater than or equal to 0. 5 of threshold value and that sensor can be isolated as the fault sensor when it＇ s two-stage decision variable appears significant jump fluctuation corresponding to other variables. This method is effective to detect and isolate vehicle active sus- pension sensor fault, to optimize suspension design, and to enhance the vehicle control reliability and use security.

关 键 词：汽车主动悬架 传感器故障 检测指标 决策函数 检测与隔离 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]