列车辅助系统传感器故障诊断与容错控制  

Sensor Fault Diagnosis and Fault-tolerant Control for Train Auxiliary Systems

作　　者：李学明 徐绍龙 成正林 陈志文[2] LI Xue-ming;XU Shao-long;CHENG Zheng-lin;CHEN Zhi-wen(Rail Transit R&D Center,Zhuzhou CRRC Times Electric Co.,Ltd.,Zhuzhou 412001,China;School of Automation,Central South University,Changsha 410083,China)

机构地区：[1]株洲中车时代电气股份有限公司轨道交通技术中心,湖南株洲412001 [2]中南大学自动化学院,湖南长沙410083

出　　处：《控制工程》2022年第8期1514-1519,1527,共7页Control Engineering of China

基　　金：国家自然科学基金资助项目(51877075,62103109)。

摘　　要：针对机车、动车组辅助系统辅变电压传感器故障后导致系统失效,无法进行容错控制的问题,提出了一种有效的故障诊断与容错控制方法。基于三相测量系统的特点,以单相二阶广义积分锁相环(SOGI-PLL)算法为基础,实现三相线电压测量值的冗余估计。利用残差生成单元和修正的贝叶斯分类算法对故障传感器进行实时检测与定位,并在传感器故障工况下用相关预估器估计值代替电压传感器信号,实现系统的容错控制。半实物仿真与现场故障案例的测试结果表明,所提算法能快速诊断出故障传感器并具有容错控制功能,可有效提升列车可靠性和运行效率,具有广阔的工程应用前景。To solve the problem of fault-tolerant control caused by the fault of auxiliary transformer voltage sensors in auxiliary systems of locomotives and electric multiple units(EMUs),an effective fault diagnosis and fault-tolerant control method is proposed in this paper.Based on the characteristics of three-phase measurement system and single-phase second order generalized integrator phase locked loop(SOGI-PLL) algorithm,redundancy estimation of the three-phase line voltage measurement value is realized.The residual generation unit and modified Bayes classification algorithm are used to detect and locate the fault sensor in real-time,and the fault-tolerant control of the system is realized by replacing the voltage sensor signal with the estimated value of the correlative estimator under the condition of sensor fault.The results of semi-physical simulation and field fault case show that the proposed algorithm can quickly diagnose the fault sensor,has fault-tolerant control function,effectively improves the reliability and efficiency of the train,and has broad engineering application prospects.

关 键 词：传感器 故障诊断 容错控制 SOGI-PLL 贝叶斯分类 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]