A novel convolutional transfer feature discrimination network for unbalanced fault diagnosis under variable rotational speeds

Abstract

Deep learning has been widely used in the field of fault diagnosis due to its excellent performance in feature extraction and has been gradually applied to solve various problems in fault diagnosis. Convolutional neural networks and transfer learning networks have been gradually employed to solve the problems of sample imbalance and domain adaptation under variable rotational speeds in fault diagnosis. However, there are still some weaknesses in current research. Firstly, sample imbalance in fault diagnosis is accompanied by the domain-adaptive problem in practice. Secondly, transfer learning cannot extract domain-invariant features that do not change with the rotational speed, which leads to poor diagnosis results when it is applied to other working conditions without transferring the rotating speed. To solve the above problems, an architecture named a convolutional transfer feature discrimination network (CTFDN) is proposed in this paper. It uses the scaled exponential linear unit activation function, and the main body adopts a two-branch network architecture of weight sharing, which mainly includes a deep feature extraction network for fusion feature extraction, a transfer learning network for domain-adaptive problems, and an unbalanced-sample feature discrimination network for feature similarity determination. Finally, two fault datasets are used to test the validity of the proposed model. The results show that the CTFDN model can extract domain-invariant features well in unbalanced fault sample information, and still has good diagnostic accuracy under variable rotational speeds.

Challenges and Opportunities of AI-Enabled Monitoring, Diagnosis & Prognosis: A Review

“…For example, Lu et al [184] and Han et al [185] used adversarial domain adaptation to train the proposed DNN to extract representative information. Xu et al [186] used adversarial domain adaptation to train a two-branch network to extract domain-invariant features, and used a scaled exponential linear unit activation function for the nonlinear activation.…”

Fusion Domain-Adaptation CNN Driven by Images and Vibration Signals for Fault Diagnosis of Gearbox Cross-Working Conditions

“…Domain adaptation techniques based on feature transfer have been much preferred in this case. One implementation of domain adaption is to add a domain adaptation term to the loss function, such as Maximum Mean Discrepancy (MMD) [ 16 , 17 , 18 ] and Wasserstein distance [ 19 ]. Another implementation of domain adaption is through domain adversarial training, in which a feature extractor aims to extract common features from both source and target domain by adversarial training [ 20 , 21 , 22 ].…”

A reinforcement transfer learning method based on a policy gradient for rolling bearing fault diagnosis

“…By transferring the knowledge learned in a certain field to similar fields, we can speed up or optimize the learning efficiency in new fields. For instance, Xu et al proposed a novel convolutional transfer feature discrimination network for imbalanced fault diagnosis at variable rotational speeds [25]. Wu et al proposed a deep transfer maximum classifier discrepancy method for rolling bearing fault diagnosis that required a small amount of labeled data [26].…”