Out-of-distribution detection-assisted trustworthy machinery fault diagnosis approach with uncertainty-aware deep ensembles

• Published in: Reliability engineering & system safety Vol. 226; p. 108648
• Main Authors: Han, Te, Li, Yan-Fu
• Format: Journal Article
• Language: English
• Published: Barking Elsevier Ltd 01.10.2022; Elsevier BV
• Subjects: Artificial neural networks; Ensemble deep learning; Fault diagnosis; Machinery; Network reliability; Neural networks; Out-of-distribution detection; Reliability engineering; Trustworthiness; Trustworthy fault diagnosis; Uncertainty; Unseen fault; Ensemble deep learning; Out-of-distribution detection; Uncertainty; Trustworthy fault diagnosis; Unseen fault
• ISSN: 0951-8320; 1879-0836
• DOI: 10.1016/j.ress.2022.108648

Recent intelligent fault diagnosis technologies can effectively identify the machinery health condition, while they are learnt based on a closed-world assumption, i.e., the training and testing data follow independently identically distribution (IID). However, in real-world diagnosis, the monitored samples are often from unknown distributions, such as unseen machine faults, leading to an out-of-distribution (OOD) problem. This is a challenging issue that may induce the model to produce unreliable and unsafe decision for unforeseen machine data. To tackle this problem, a novel OOD detection-assisted trustworthy machinery fault diagnosis approach is developed to enhance the reliability and safety of intelligent models. First, multiple deep neural networks are integrated to establish an ensemble diagnosis system, called deep ensembles. Then, the trustworthy analysis with uncertainty-aware deep ensembles is conducted to detect the OOD samples and issue the warnings for the potential untrustworthy diagnosis. A selection criterion of uncertainty threshold is given. Finally, the trustworthy decisions are achieved by comprehensively considering the deep ensembles’ prediction and uncertainty. The proposed trustworthy fault diagnosis approach is validated in two case studies, exhibiting significant advantages for diagnosing OOD samples. •Develop an out-of-distribution detection-assisted trustworthy fault diagnosis method.•Utilize the ensemble of deep neural networks to quantify uncertainty during decision.•Uncertainty can increase the transparency of intelligent fault diagnosis model.•Foster the collaboration between intelligent diagnosis model and human.