Deep Learning -Based Remaining Life Prediction for DC-Link Capacitor in High Speed Train
ID：128
Submission ID：48 View Protection：ATTENDEE
Updated Time：2023-11-20 13:53:19 Hits：156
Poster Presentation
Start Time：Pending （Asia/Shanghai）
Duration：Pending
Session：[No Session] » [No Session Block]
Presentation File
Tips: The file permissions under this presentation are only for participants. You have not logged in yet and cannot view it temporarily.
Abstract
The accurate prediction of the remaining useful life of DC-link capacitors is crucial in high-speed railway traction drive operating conditions. This prediction takes into account environmental factors, physical structure, and strict safety operation requirements. In this paper, we propose a method for predicting remaining life using deep learning optimized by the particle swarm algorithm. Firstly, the upper and lower voltage of the capacitor are selected as the characteristic values. By denoising the voltage with wavelet transform, we calculate failure thresholds for both ends of the support capacitor. Then, we amplify the input weights in real-time using the Long Short-Term Memory (LSTM) neural network with a macro-micro attention mechanism. Finally, we utilize the particle swarm optimization algorithm to optimize the number of input units and the learning rate of the neural network for the purpose of predicting lifetime. The effectiveness of this method is verified through model evaluation indices in a case study of high speed train traction system.
Keywords
Residual life prediction, support capacitors, long short-term memory neural networks, macro-micro attention mechanism, particle swarm optimization.
Speaker
Comment submit