The railway turnout fault diagnosis algorithm based on BP neural network

This paper presents an intelligent detection algorithm based on BP Neural Network, which is based on the current curve change rule of the turnout switch machine. Firstly it analyzes characteristics of each stage of turnouts device operating current curve, summarizes the typical turnout fault operating current curve; Then, establishes the mapping data sets between the action current and turnout fault types; Finally, using the BP neural network to train and test the mapping data sets of action current and turnout fault types. Experimental results show that the algorithm has better adaptability, high accuracy, easy installation and low cost, and does not involve the station interlocking equipment when it is upgraded.