Prediction Method for Rolling Bearing RUL Based on Hierarchical Adaptive Wavelet Threshold Denoising and CNN-LSTM

Aiming at the low accuracy of Remaining Useful Life (RUL) prediction caused by the difficulty of fault feature extraction of rolling bearing vibration signals in a strong noise environment, this paper proposes an RUL prediction method based on the combination of Hierarchical Adaptive Wavelet Thresholding (HAWT) and CNN-LSTM for RUL prediction. The method first adds high-intensity Gaussian white noise to the original signal, then denoises it using HAWT, and finally constructs a CNN-LSTM model to predict the lifetime. Experiments with the XJTU-SY rolling bearing accelerated life dataset show that the method can significantly improve the accuracy of rolling bearing RUL prediction compared with other methods, providing a reliable basis for rolling bearing health management.