In high-speed railway traction power supply systems, effective intelligent fault diagnosis measures assist operation and maintenance staff in swiftly addressing faults, promptly restoring power supply, or minimizing the power outage area. After accumulating a substantial amount of historical data, monitoring systems can discern the operational and transitional trends of equipment. By integrating current structural characteristics, parameters, and environmental conditions, it is possible to predict the future status of equipment, foreseeing the likelihood of imminent faults. This provides a reliable basis for early fault detection and prevention of fault-related losses. Predictive science is an emerging discipline where prognostics involves diagnosing parts or system functional status predictively, based on current or historical performance. Guided by relevant theories and methods, it analyzes and infers the future development status and trends of the study object. Prognostics technologies are currently widely applied in industries such as industrial, commercial, financial, and meteorological fields. In recent years, researchers at home and abroad have utilized methods such as artificial neural networks, support vector machines, Bayesian networks, and hidden Markov models in the field of power system prognostics. With the continuous development of high-tech, traction power supply systems and large industrial equipment are becoming increasingly complex, with higher levels of automation and integration. This complexity in the causes and signs of faults poses greater challenges for prognostics, necessitating the integration of various prognostics techniques and the application of new prognostics theories and classification methods. This promotes the co-development of prognostics, state management technology, and fault diagnosis.