Enhancing Fault Detection in High-Speed Train Dampers with Auxiliary Task Learning

The damper in the suspension system of a high-speed train is an important component, and the failure and performance degradation of the damper will affect the dynamic behaviors of the train, so it plays a crucial role in the safety of the train operation. In this paper, we propose an auxiliary task learning method based on multi-task learning for fault detection in high-speed train dampers. By taking the problem of estimating the performance degradation of the damper as an auxiliary task and combining uncertainty to weight losses, the accuracy of fault detection in the damper is improved. Experimental results show that the proposed method achieves better performance in fault detection of dampers.