Algorithm of Railway Turnout Fault Detection Based on PNN Neural Network
Kai Zhang, Kai Du, Yongfeng Ju · 2014
This paper presents a turnout fault detection algorithm based on PNN neural network. This algorithm summarized the typical turnout fault action current curves, established the mapping data sets between the action current curve and turnout fault types, used PNN neural network and BP neural network to train and test the mapping data sets of action current curve. Experimental results show that the turnout fault detection algorithm based on PNN neural network is better than BP neural network algorithm. It has higher precision and less parameter adjustment, easy to set up and so on.