Research on ZPW-2000R Track Circuit Fault Diagnosis Based on Neural Network. Issue 1 (March 2021)

• Record Type: Journal Article
• Authors: Lu, Jiao; Yu, Jianli; Huang, Chunlei; Chen, Honggen
• Is Part Of: IOP conference series. Volume 693:Issue 1(2021)
• Publication Date: 2021-03
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/693/1/012077
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: In view of the situation that the fault monitoring is not timely and the accuracy of fault diagnosis is low due to only indoor monitoring equipment but not outdoor monitoring equipment, an intelligent diagnosis method based on neural network is proposed. Firstly, according to the different operating environment and influencing factors of the track circuit, the method of orthogonal test is designed to obtain the information of each component There are three kinds of track circuit monitoring variables with complex environmental factors, and then the track circuit fault types are divided into three categories: sending channel fault, receiving channel fault and track fault. On the basis of rapid location of track circuit fault location, 21 kinds of specific fault types are distinguished. The simulation results show that: when using this method for track circuit fault diagnosis, even without outdoor monitoring data, it can quickly locate the location of the fault, and the location accuracy can be up to 100%. When identifying 21 fault types, the accuracy of fault diagnosis can still be up to 90%, the diagnosis rate is high and the judgment is fast, which can assist the field maintenance personnel judge faults accurately and quickly.