PHM技术在高速动车组关键部件检修中的应用

Application of PHM Technology in the Maintenance of Key Components of High-Speed Rolling Stock

为提升高速动车组运行的安全性与稳定性,分析动车组部件结构特点,基于PHM技术,研究PHM技术在高速动车组关键部件检修中的应用。通过车载通信模块采集动车组上各系统内传感器数据,通过网络实时回传地面,构成了动车组PHM的数据监测单元。应用人工智能技术,提取并处理故障特征数据,建立PHM的故障特征数据单元,为动车组部件的检修提供数据基础,利用BP神经网络算法,完成故障的智能诊断;设定合理的检修方案,构成PHM的故障诊断与检修单元,完成高速动车组关键部件检修。为检测研究方法的检修效果,设计模拟实验,结果显示,研究方法对100组故障数据信息节点的平均处理时间比对照组快1.44 s,对不同类型的故障信息检测的准确率均高于对照组,分别高7.39%、4.67%、1.67%、8.92%,具有高效性与准确性,同时能够提高列车运行的安全可靠性。

In order to improve the safety and stability of high-speed rolling stock operation,the structural characteristics of rolling stock components are analyzed and the application of PHM technology in the maintenance of key components of high-speed rolling stock is studied based on PHM technology.The data of sensors in each system on the rolling stock is collected through the on-board communication module and transmitted back to the ground in real time through the network,which constitutes the data monitoring unit of the PHM of the rolling stock.Artificial intelligence technology is applied to extract and process fault feature data,establish the fault feature data unit of PHM,provide data basis for the overhaul of rolling stock components,use BP neural network algorithm to complete the intelligent diagnosis of faults;set a reasonable overhaul plan,constitute the fault diagnosis and overhaul unit of PHM,and complete the overhaul of key components of high-speed rolling stock.In order to test the overhaul effect of the research method,design simulation experiments,results show that the research method,the average processing time of 100 groups of fault data information nodes is 1.44 s faster than the control group,and the accuracy rate of different types of fault information detection are higher than the control group,7.39%,4.67%,1.67%,8.92%respectively,with high efficiency and accuracy,and at the same time can improve It is efficient and accurate,and can improve the safety and reliability of train operation.