Abstract:Recently, deep learning-based intelligent fault diagnosis techniques have obtained good classification performance with amount of supervised training data. However, domain shift problem between the training and testing data usually occurs due to variation in operating conditions and interferences of environment noise. Transfer learning provides a promising tool for handling the cross-domain diagnosis problems by leveraging knowledge from the source domain to help learning in the target domain. Most existing studies attempt to learn both domain features in a common feature space to reduce the domain shift, which are not optimal on specific discriminative tasks and can be limited to small shifts. This article proposes a novel domain adversarial transfer network (DATN), exploiting task-specific feature learning networks and domain adversarial training techniques for handling large distribution discrepancy across domains. First, two asymmetric encoder networks integrating deep convolutional neural networks are designed for learning hierarchical representations from the source domain and target domain. Then, the network weights learned in source tasks are transferred to improve training on target tasks. Finally, domain adversarial training with inverted label loss is introduced to minimize the difference between source and target distributions. To validate the effectiveness and superiority of the proposed method in the presence of large domain shifts, two fault data sets from different test rigs are investigated, and different fault severities, compound faults, and data contaminated by noise are considered. The experimental results demonstrate that the proposed method achieves the average accuracy of 96.45% on the bearing data set and 98.92% on the gearbox data set, which outperforms other algorithms.

Multisource Domain Transfer Learning for Bearing Fault Diagnosis

Moment matching-based intraclass multisource domain adaptation network for bearing fault diagnosis

A New Deep Transfer Learning Method for Bearing Fault Diagnosis under Different Working Conditions

A Fault Diagnosis Method for Rolling Bearing Based on Deep Adversarial Transfer Learning with Transferability Measurement

A Sub-Domain Adaptive Transfer Learning Base on Residual Network for Bearing Fault Diagnosis

A Fault Diagnosis Method of Bearings Based on Deep Transfer Learning.

A multi-source information transfer learning method with subdomain adaptation for cross-domain fault diagnosis

Domain Adversarial Transfer Network for Cross-Domain Fault Diagnosis of Rotary Machinery

Multisource cross-domain fault diagnosis of rolling bearing based on subdomain adaptation network

Deep Transfer Learning for Bearing Fault Diagnosis: A Systematic Review Since 2016

Bearing Fault Diagnosis Based on Multiple Feature Transfer Learning Network

Conditional Distribution-Guided Adversarial Transfer Learning Network with Multi-Source Domains for Rolling Bearing Fault Diagnosis

Distance-guided domain adaptation for bearing fault diagnosis under variable operating conditions

Residual deep subdomain adaptation network: A new method for intelligent fault diagnosis of bearings across multiple domains

A New Bearing Fault Diagnosis Method Based on Deep Transfer Network and Supervised Joint Matching

Bearing Fault Diagnosis Based on Local Manifold Discriminant Domain Adaptation

Multi-scale deep intra-class transfer learning for bearing fault diagnosis

A Hybrid Transfer Learning Method For Fault Diagnosis Of Machinery Under Variable Operating Conditions

Fault diagnosis of rolling bearing based on cross-domain divergence alignment and intra-domain distribution alienation

Semi-Supervised Transfer Learning Method for Bearing Fault Diagnosis with Imbalanced Data