Fault Diagnosis of Jointless Track Circuit Based on ReliefF-C4.5 Decision Tree. Issue 1 (1st December 2022)

• Record Type: Journal Article
• Authors: Liu, Ruidan; Dong, Yu
• Is Part Of: Journal of physics. Volume 2383 Issue 1(2022)
• Publication Date: 2022-12-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2383/1/012047
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract : At this stage, when judging the fault type of jointless track circuit, it mainly relies on the staff to analyze the computer monitoring data. This fault identification method has problems such as high dependence on staff and long fault identification cycle. In order to solve the above problems, the combination of the reliefF algorithm and the C4.5 decision tree is introduced into the fault diagnosis of the jointless track circuit. First, the reliefF algorithm is used to give weights to the selected 9 feature parameters according to the fault classification ability of the features. Second, sort the feature parameters according to the weight, and select the first seven feature parameters as the optimal feature subset. Finally, complete the training and testing of the classifier with the optimal feature subset. The experimental results show that the fault diagnosis accuracy of ReliefF-C4.5 Decision Tree is 93.33%, So the fault diagnosis model proposed in this paper is suitable for fault diagnosis of track circuit.