Prediction of the Remaining Useful Life for the Power Module in the Traction System of Maglev Trains
ID：244
Submission ID：244 View Protection：PUBLIC
Updated Time：2021-06-18 18:15:57 Hits：514
Poster Presentation
Abstract
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.
Keywords
Power Module， Traction System， Remaining Useful Life， Dynamic Time Warping， Kernel Density Estimator
Submission Author
Biao Yang
National University of Defense Technology
Comment submit