A method for fault detection in multi-component systems based on sparse autoencoder-based deep neural networks. (April 2022)

• Record Type: Journal Article
• Authors: Yang, Zhe; Baraldi, Piero; Zio, Enrico
• Is Part Of: Reliability engineering & system safety. Volume 220(2022)
• Publication Date: 2022-04
• Subjects: Deep learning -- Sparse autoencoder -- Deep neural network -- Fault detection -- Evolving environment -- Multi-component system; 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.2021.108278
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: We consider the problem of detecting failures in multi-component system. The autoencoder-based method is able to automatically extract degradation indicators. A single run-to-failure trajectory is enough to pre-train the deep neural network. The computational burden of deep neural network hyperparameter setting is reduced. The proposed method allows dealing with unbalanced dataset. Abstract: In multi-component systems, degradation, maintenance, renewal and operational mode change continuously the operating conditions. The identification of the onset of abnormal conditions from signal measurements taken in such evolving environments can be quite challenging, due to the difficulty of distinguishing the real cause of the signal variations. In this work, we present a method for fault detection in evolving environments that uses a Sparse Autoencoder-based Deep Neural Network (SAE-DNN) and a novel procedure that remarkably reduces the computational burden for setting the values of the hyperparameters. The method is applied to a synthetic case study and to a bearing vibration dataset. The results show that it is able to accurately detect faults in multi-component systems, outperforming other state-of-the-art methods.