Adaptive deep learning-based remaining useful life prediction framework for systems with multiple failure patterns. (July 2023)

• Record Type: Journal Article
• Authors: Xiong, Jiawei; Zhou, Jian; Ma, Yizhong; Zhang, Fengxia; Lin, Chenglong
• Is Part Of: Reliability engineering & system safety. Volume 235(2023)
• Publication Date: 2023-07
• Subjects: Remaining useful life -- Failure mode recognition -- Deep learning -- Prediction model; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2023.109244
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Propose a novel adaptive framework to predict RUL under multiple failure modes. Design a DCNN-based failure mode recognizer incorporating physics-informed classifier. Adaptively train and apply LSTM based prediction models based on failure modes. Operation condition-based Savitzky-Golay filter improves the RUL prediction accuracy. Validate the performance gains of the proposed framework compared with ten methods. Abstract: Recent advances in multivariate data fusion technology have promoted the applications of neural network-based models for remaining useful life (RUL) prediction. However, the interpretability of these models is usually poor since they are developed in a black-box manner. It is difficult to use them in engineering systems with multiple failure modes (FMs) under various operation conditions (OCs). This work proposes an adaptive deep learning-based RUL prediction framework with FM recognition. First, a FM recognizer fusing physics-informed FM classifier with deep convolutional neural networks (DCNN) is developed, which improves the interpretability and the accuracy of the recognition model. Then, a framework which can adaptively train models and select them for RUL prediction according to FM recognition results is presented. An OC-based smoothing technique is proposed to improve the RUL prediction accuracy and robustness. Extensive experiments based on turbofan datasets are conducted to validate the effectiveness of the proposed framework. The results show that the RUL prediction accuracy is improved by 7% under the proposed framework when compared with other methods. It proves the performance gains of the proposed framework by incorporating prior FM recognition with RUL prediction. It also provides insights for RUL prognostics subject to distinct FMs and OCs.