A Robust and Sustainable Approach to Pantograph–Catenary Fault Diagnosis Using Sta and Data-Driven Methods

Abstract

The pantograph-catenary system is critical in maintaining a stable power supply for high-speed railways. However, contact instability and component wear can lead to degradation and unexpected failures. This paper proposes an integrated framework that combines the Leveraging Super Spiral Algorithm (STA) and machine learning (ML) for robust and sustainable fault diagnosis in pantograph-catenary systems. The STA ensures optimal contact force under dynamic operating conditions, while a vibration analysis unit captures real-time operational data. Time-frequency features extracted from these signals train ML models to predict the remaining useful life (RUL) and detect anomalies early. Additionally, a real-time warning system enables proactive maintenance planning. Simulations and experimental results demonstrate that this framework significantly improves contact stability, fault detection accuracy, and maintenance efficiency compared to traditional threshold-based methods, thereby facilitating the implementation of intelligent maintenance strategies in future high-speed railways.