Su, Baoyi (Beijing Jiaotong University), Dai, Shenghua (Beijing Jiaotong University)

Research on Switch Fault Prediction Based on Sparrow Search Algorithm Optimization Extreme Gradient Boosting

Scheduled for presentation during the Poster Session "Data Mining and Data Analysis" (FrAT14), Friday, September 27, 2024, 10:30−12:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on October 8, 2024

Abstract

In order to further improve the maintenance efficiency of railway signaling equipment and ensure its safety and reliability, based on the switch action current curve data of the microcomputer monitoring system of the electric section of the railway bureau, a switch fault prediction method based on the Sparrow algorithm optimized XGBoost was adopted. Firstly,according to the 7 common fault modes and 1 normal mode on site, the microcomputer monitoring fault data of ZD6 switch equipment was collected in chronological order. Secondly, a fault prediction model of switch equipment based on Sparrow algorithm optimization XGBoost was constructed, because the hyperparameter of the XGBoost algorithm directly affect the performance of the prediction model, the Sparrow algorithm is used to optimize the hyperparameters of the switch equipment fault prediction model to deeply mine the switch before the fault occurs. Finaly, experimental verification: this method can accurately predict the fault status of switch equipment, so that site maintenance personnel can predict the fault in advance and take maintenance measures, improve the maintenance efficiency of switch equipment and ensure the safe operation of trains. This study provides a new method for research related to fault prediction of switch equipment.