Research on Predictive Maintenance of Traction Converter Filter Based on PHM System

Abstract

Based on the existing PHM system, the predictive maintenance research for the traction converter filter of the standard EMU train is studied. This paper studied the influence of filter plugging on the performance of cooling system, and established a predictive model for traction converter filter maintenance through big data analysis and machine learning methods. As a result, it can guide the filter maintenance, reduce the cost of maintenance and the workload of personnel, ensure the safety of the train operation, and make EMU operations shift from traditional planned maintenance to predictive maintenance based on condition.