基于神经网络和马尔科夫模型的道岔故障诊断与预测研究  

作　　者：陈红霞[1] 孔筱筱 

机构地区：[1]南京铁道职业技术学院,南京210031

出　　处：《科技创新与应用》2023年第33期98-101,107,共5页Technology Innovation and Application

基　　金：江苏省高铁安全工程技术研究开发中心开放基金资助(GTAQ2021008);南京铁道职业技术学院科研基金项目(Yz210005)。

摘　　要：该文基于神经网络和马尔科夫模型进行道岔的故障诊断与预测研究,首先对道岔的动作过程进行分析,通过对采集的非故障功率数据进行比较分析,退化状态是故障诊断到故障预测的重要过程,通过分析退化状态的形成原因,进一步分析退化状态数据与现场出现的典型故障状态数据之间的关系,通过隐马尔科夫模型建立退化状态的典型模型,用于退化状态的识别以及故障状态的评估,实现道岔故障状态的诊断和道岔故障的预测,同时也设计道岔故障诊断和预测的系统平台,该研究对于现场道岔故障的诊断和预测有很好的参考价值。This paper studies the fault diagnosis and prediction of switch based on neural network and Markov model.Firstly,the operation process of switch is analyzed.By comparing and analyzing the collected non-fault power data,the degraded state is an important process from fault diagnosis to fault prediction.The relationship between the degraded state data and the typical fault state data in the field is further analyzed.The typical model of the degraded state is established through the hidden Markov model,which is used to identify the degraded state and evaluate the fault state,so as to realize the diagnosis and prediction of the fault state of the switch.At the same time,the system platform of the fault diagnosis and prediction of the switch is also designed.The research has a good reference value for the diagnosis and prediction of switch faults.

关 键 词：神经网络 马尔科夫模型 道岔故障 退化状态 高速铁路 

分 类 号：U284.5[交通运输工程—交通信息工程及控制]