Imbalanced Fault Diagnosis of Rolling Bearing Using Data Synthesis Based on Multi-Resolution Fusion Generative Adversarial Networks

被引:14
作者
Hao, Chuanzhu [1 ]
Du, Junrong [2 ,3 ]
Liang, Haoran [2 ,3 ]
机构
[1] Shandong Huayu Univ Technol, Coll Elect Engn, Dezhou 253034, Peoples R China
[2] Chinese Acad Sci, Key Lab Space Utilizat, Technol & Engn Ctr Space Utilizat, Beijing 100094, Peoples R China
[3] Univ Chinese Acad Sci, Beijing 100049, Peoples R China
基金
中国国家自然科学基金;
关键词
imbalanced fault diagnosis; data augmentation; generative adversarial networks; feature transfer; CLASSIFICATION; SMOTE;
D O I
10.3390/machines10050295
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Fault diagnosis of industrial bearings plays an invaluable role in the health monitoring of rotating machinery. In practice, there is far more normal data than faulty data, so the data usually exhibit a highly skewed class distribution. Algorithms developed using unbalanced datasets will suffer from severe model bias, reducing the accuracy and stability of the classification algorithm. To address these issues, a novel Multi-resolution Fusion Generative Adversarial Network (MFGAN) is proposed for the imbalanced fault diagnosis of rolling bearings via data augmentation. In the data-generation process, the improved feature transfer-based generator receives normal data as input to better learn the fault features, mapping the normal data into fault data space instead of random data space. A multi-scale ensemble discriminator architecture is designed to replace original single discriminator structure in the discriminative process, and multi-scale features are learned via ensemble discriminators. Finally, the proposed framework is validated on the public bearing dataset from Case Western Reserve University (CWRU), and experimental results show the superiority of our method.
引用
收藏
页数:18
相关论文
共 41 条
[1]  
[Anonymous], 2003, WORKSHOP LEARNING IM
[2]  
[Anonymous], 2018, P INT C LEARN REPR
[3]   SMOTE: Synthetic minority over-sampling technique [J].
Chawla, Nitesh V. ;
Bowyer, Kevin W. ;
Hall, Lawrence O. ;
Kegelmeyer, W. Philip .
2002, American Association for Artificial Intelligence (16)
[4]   RSMOTE: A self-adaptive robust SMOTE for imbalanced problems with label noise [J].
Chen, Baiyun ;
Xia, Shuyin ;
Chen, Zizhong ;
Wang, Binggui ;
Wang, Guoyin .
INFORMATION SCIENCES, 2021, 553 :397-428
[5]   A Review of Fault Detection and Diagnosis for the Traction System in High-Speed Trains [J].
Chen, Hongtian ;
Jiang, Bin .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2020, 21 (02) :450-465
[6]   Cost-Sensitive Large margin Distribution Machine for classification of imbalanced data [J].
Cheng, Fanyong ;
Zhang, Jing ;
Wen, Cuihong .
PATTERN RECOGNITION LETTERS, 2016, 80 :107-112
[7]   Near-Bayesian Support Vector Machines for imbalanced data classification with equal or unequal misclassification costs [J].
Datta, Shounak ;
Das, Swagatam .
NEURAL NETWORKS, 2015, 70 :39-52
[8]   Motion Classification Using Kinematically Sifted ACGAN-Synthesized Radar Micro-Doppler Signatures [J].
Erol, Baris ;
Gurbuz, Sevgi Zubyede ;
Amin, Moeness G. .
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS, 2020, 56 (04) :3197-3213
[9]  
Esteban Cristobal, 2017, arXiv
[10]   Using Cost-Sensitive Learning and Feature Selection Algorithms to Improve the Performance of Imbalanced Classification [J].
Feng, Fang ;
Li, Kuan-Ching ;
Shen, Jun ;
Zhou, Qingguo ;
Yang, Xuhui .
IEEE ACCESS, 2020, 8 :69979-69996