Intelligent prediction of rolling bearing remaining useful life based on probabilistic DeepAR-Transformer model

Abstract

Remaining useful life (RUL) prediction for rolling bearings requires highly accurate and stable long-term prediction capabilities in equipment health management, which demands that the prediction model has strong data reasoning and regression performance. However, it is difficult to accurately capture long-term dependencies via traditional convolutional neural network because the information loss and insufficient analysis are unavoidable during the feature extraction process. An end-to-end time series forecasting method called D-former for RUL prediction of rolling bearings is proposed in this paper. The method mainly consists of DeepAR and a multi-layer encoder, so it is able to extract degradation features directly from the original signal. This method has the following salient features: (1) the designed multi-head attention mechanism can highlight important feature information and realize parallel computing, so the method is extremely suitable for processing long-term time series; (2) the important time feature information is rearranged through DeepAR, so the method has the excellent domain adaptability, and it can achieve accurate prediction of rolling bearing RUL under different working conditions. The verification experiment was implemented on the IEEE PHM 2012 dataset and the XJTU-SY bearing dataset. The experimental results show that the proposed D-former method is actually superior to the existing mainstream RUL prediction methods.

Remaining useful life prediction method based on stacked autoencoder and generalized wiener process for degrading bearing

“…To illustrate the effectiveness of the suggested approach, a comparative analysis was conducted using methods from the following references to predict the RUL of the studied bearings. Specifically, the model-based method from [34] is referred to as M2, [35] is referred to as M3, [36] is referred to as M4, and [37] is referred to as M5. Additionally, the performance of the prediction methods was comprehensively evaluated using the scoring function from the IEEE PHM 2012 Challenge [38].…”

A lightweight transformer and depthwise separable convolution model for remaining useful life prediction of turbofan engines

“…In contrast, deep learning-based methods have emerged as the mainstream approach, which is also the methodology employed in this study. The commonly used methods include convolutional neural network (CNN) [20], autoencoder (AE) [21], recurrent neural network (RNN) and variant models of RNN [22], and transformer [23]. In general, the failure of mechanical equipment is typically a process that evolves over time, transitioning gradually from an initial healthy state to a faulty state, resulting in interdependent patterns of failurerelated information in the monitoring data.…”