TY - JOUR

T1 - Adversarial domain-invariant generalization

T2 - a generic domain-regressive framework for bearing fault diagnosis under unseen conditions

AU - Chen, L.

AU - Li, Q.

AU - Shen, C.

AU - Zhu, J.

AU - Wang, D.

AU - Xia, M.

N1 - ©2021 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

PY - 2022/3/31

Y1 - 2022/3/31

N2 - Recently, various fault diagnosis methods based on domain adaptation (DA) have been explored to solve the problem of discrepancy between the source and target domains. However, given complex industrial scenarios, DA-based methods usually fail when the working conditions of machines are unseen, i.e., target data are unavailable during model training. In this work, a generic domain-regressive framework for fault diagnosis, namely, adversarial domain-invariant generalization (ADIG), is proposed. ADIG leverages multiple available domain data to exploit domain-invariant knowledge through adversarial learning between the feature extractor and the domain classifier. Simultaneously, the fault classifier generalizes the knowledge from the source-related domain to diagnose the unseen but related target domain signals. Moreover, customized strategies of feature normalization and adaptive weight are proposed to promote diagnosis performance. Comprehensive case studies show that ADIG achieves satisfactory diagnosis accuracy and robustness under unseen conditions, indicating that ADIG is a remarkably potential diagnosis tool for real-case industrial machines. IEEE

AB - Recently, various fault diagnosis methods based on domain adaptation (DA) have been explored to solve the problem of discrepancy between the source and target domains. However, given complex industrial scenarios, DA-based methods usually fail when the working conditions of machines are unseen, i.e., target data are unavailable during model training. In this work, a generic domain-regressive framework for fault diagnosis, namely, adversarial domain-invariant generalization (ADIG), is proposed. ADIG leverages multiple available domain data to exploit domain-invariant knowledge through adversarial learning between the feature extractor and the domain classifier. Simultaneously, the fault classifier generalizes the knowledge from the source-related domain to diagnose the unseen but related target domain signals. Moreover, customized strategies of feature normalization and adaptive weight are proposed to promote diagnosis performance. Comprehensive case studies show that ADIG achieves satisfactory diagnosis accuracy and robustness under unseen conditions, indicating that ADIG is a remarkably potential diagnosis tool for real-case industrial machines. IEEE

KW - Adaptation models

KW - Adversarial learning

KW - Cross-domain fault diagnosis

KW - Data models

KW - Domain generalization

KW - Fault diagnosis

KW - Feature extraction

KW - Generative adversarial networks

KW - Rotating machinery

KW - Task analysis

KW - Training

KW - Failure analysis

KW - Bearing fault diagnosis

KW - Diagnosis performance

KW - Fault diagnosis method

KW - Feature extractor

KW - Feature normalization

KW - Industrial machines

KW - Industrial scenarios

KW - Fault detection

U2 - 10.1109/TII.2021.3078712

DO - 10.1109/TII.2021.3078712

M3 - Journal article

VL - 18

SP - 1790

EP - 1800

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

SN - 1551-3203

IS - 3

ER -