Prediction of the Remaining Useful Life for the Power Module in the Traction System of Maglev Trains

Model-based prediction methods are difficult to capture the physical process of system degradation, and although artificial intelligence-based prediction methods do not require much prior knowledge, it is difficult to pass existing data due to the lack of operational data for Power Module in the Traction System of Maglev Trains Before forecasting, find an appropriate model to predict the future development of degradation indicators. In this regard, based on health assessment, combined with Dynamic Time Warping (DTW) and Kernel Density Estimator (KDE), an improved similarity remaining life prediction method was studied.