Optimal transport strategy-based meta-attention network for fault diagnosis of rotating machinery with zero sample
Authors: 
Ke Wu, Kaiwei Yu, Chong Chen, Jun Wu, Yan LiuAuthors Info & Claims
Ke Wu, Kaiwei Yu, Chong Chen, Jun Wu, Yan LiuAuthors Info & ClaimsPages 6799 - 6815
Abstract
Deep learning-based methods are widely applied to fault diagnostics, which depend on adequate samples. However, fault samples are often limited and even not available in industrial applications. To solve this problem, an optimal transport strategy-based meta-attention network (OTS-MAN) is proposed for the fault diagnosis by exploiting the fault knowledge learned from few source domains to diagnose the target domain with zero sample. Firstly, a new meta-attention network is built to mine discriminative features of each class from the source domain. Then, an optimal transport strategy is designed to align the feature distribution of each category between known fault in the source domain and unknown fault in the target domain. Finally, the similarity scores are obtained to assess the health status of the target domain. The proposed OTS-MAN is trained only with known source domain data and can diagnose unknown faults without previous access to target domain data. The validity of the proposed method is implemented through using two cases. The results indicate that the OTS-MAN has a better fault diagnosis accuracy than existing methods, and its noise immunity is also improved.
References
[1]
Xing ZY, Zhao RZ, Wu YC, and He TJ Intelligent fault diagnosis of rolling bearing based on novel CNN model considering data imbalance Appl Intell 2022 52 16281-16293
[2]
Hou YD, Ma JL, Wang JJ, Li TZ, and Chen ZQ Enhanced generative adversarial networks for bearing imbalanced fault diagnosis of rotating machinery Appl Intell 2023 53 21 25201-15
[3]
Meng Z, He HH, Cao W, Li JM, Cao LX, Fan JJ, Zhu M, and Fan FJ A novel generation network using feature fusion and guided adversarial learning for fault diagnosis of rotating machinery Expert Syst Appl 2023 234 121058
[4]
Zhang ZQ, Zhou FN, Karimi HR, Fujita H, Hu X, Wen CL, and Wang TZ Attention gate guided multiscale recursive fusion strategy for deep neural network-based fault diagnosis Eng Appl Artif Intel 2023 126 107052
[5]
Lu N, Cui ZY, Hu HY, and Yin T Multi-view and Multi-level network for fault diagnosis accommodating feature transferability Expert Syst Appl 2023 213 119057
[6]
Su J, Song HJ, and Song FY Fault diagnosis of steam power system based on convolutional neural network, Chinese J Ship Res 2022 17 96-102
[7]
Xu J, Zhang H, Zhou L, and Fan Y Zero-Shot Compound Fault Diagnosis Method Based on Semantic Learning and Discriminative Features IEEE Instrum Meas 2023 72 1-13
[8]
Chen ZY, Wu J, Deng C, Wang XQ, and Wang YH Deep Attention Relation Network: A Zero-Shot Learning Method for Bearing Fault Diagnosis Under Unknown Domains IEEE T Reliab 2022 72 1 79-89
[9]
Pan TY, Chen JL, Xie JS, Zhou ZT, and He SL Deep Feature Generating Network: A New Method for Intelligent Fault Detection of Mechanical Systems Under Class Imbalance IEEE T Ind Inform 2021 17 6282-6293
[10]
Xu K, Kong XG, Wang QB, Yang SK, Huang NN, and Wang JJ A bearing fault diagnosis method without fault data in new working condition combined dynamic model with deep learning Adv Eng Inform 2022 54 101795
[11]
Zhou FN, Yang S, Fujita H, Chen DM, and Wen CL Deep learning fault diagnosis method based on global optimization GAN for unbalanced data Knowl-Based Syst 2020 187 104837
[12]
Yu JB and Liu GL Knowledge extraction and insertion to deep belief network for gearbox fault diagnosis Knowl-Based Syst 2020 197 105883
[13]
Gao DW, Zhu YS, Ren ZJ, Yan K, and Kang W A novel weak fault diagnosis method for rolling bearings based on LSTM considering quasi-periodicity Knowl-Based Syst 2021 231 107413
[14]
Wu YC, Zhao RZ, Jin WY, He TJ, Ma SC, and Shi MK Intelligent fault diagnosis of rolling bearings using a semi-supervised convolutional neural network Appl Intell 2021 51 2144-2160
[15]
Chen ZY, Wu J, Deng C, Wang C, and Wang YH Residual deep subdomain adaptation network: A new method for intelligent fault diagnosis of bearings across multiple domains Mech Mach Theory 2022 169 104635
[16]
Liu JF, Yu X, and Wan HB Rolling bearing fault diagnosis method based on modified fourier mode decomposition and band entropy Chin J Ship Res 2022 17 190-197
[17]
Zheng M, Zhi K, Zeng J, Tian C, and You L A hybrid CNN for image denoising J Artificial Intell Technol 2022 2 93-99
[18]
Zhou F, Liu S, Fujita H, Hu X, Zhang Y, Wang B, and Wang K Fault diagnosis based on federated learning driven by dynamic expansion for model layers of imbalanced client Expert Syst Appl 2024 238 121982
[19]
Zhang HD, Yuan X, Li DY, You J, Liu B, Zhao XM, Cai WM, and Ju S An effective framework using identification and image reconstruction algorithm for train component defect detection Appl Intell 2022 52 10116-10134
[20]
Fan WT, Huang HY, Liang C, Liu X, and Peng SJ Unsupervised meta-learning via spherical latent representations and dual VAE-GAN Appl Intell 2023 53 19 22775-88
[21]
Wang X, Jiang HK, Liu YP, and Yang Q Data-augmented patch variational autoencoding generative adversarial networks for rolling bearing fault diagnosis Meas Sci Technol 2023 34 5 055102
[22]
Yang S, Zhou YF, Chen X, Deng CY, and Li C Fault diagnosis of wind turbines with generative adversarial network-based oversampling method Meas Sci Technol 2023 34 4 044004
[23]
Hu WY, Wang TY, and Chu FL A Wasserstein generative digital twin model in health monitoring of rotating machines Comput Ind 2023 145 103807
[24]
Su CW, Wang XR, Liu RJ, Guo ZY, Sang ST, Yu S, and Zhang HF Fault diagnosis method based on triple generative adversarial nets for imbalanced data Meas Sci Technol 2023 34 3 035007
[25]
Cabrera D, Villacis M, Cerrada M, Sanchez RV, Li C, Sancho F, Long JY, and Estupinan E Adversarial Fault Detector Guided by One-Class Learning for a Multistage Centrifugal Pump, IEEE-Asme T Mech 2022
[26]
Shang ZW, Zhang J, Li WX, Qian SQ, Liu JY, and Gao MS A Novel Small Samples Fault Diagnosis Method Based on the Self-attention Wasserstein Generative Adversarial Network, Neural Process Lett 2023
[27]
Tang Z, Bo L, Liu XF, and Wei DP A semi-supervised transferable LSTM with feature evaluation for fault diagnosis of rotating machinery Appl Intell 2022 52 1703-1717
[28]
Xue XA, Zhao XP, Zhang YH, Ma MY, Bu C, and Peng P Federated transfer learning with consensus knowledge distillation for intelligent fault diagnosis under data privacy preserving Meas Sci Technol 2024 35 1 015108
[29]
Tan HC, Xie SC, Ma W, Yang CX, and Zheng SW Correlation feature distribution matching for fault diagnosis of machines Reliab Eng Syst Safe 2023 231 108981
[30]
Mao WT, Wang GS, Kou LL, and Liang XH Deep Domain-Adversarial Anomaly Detection With One-Class Transfer Learning IEEE-Caa J Automatic 2023 10 524-546
[31]
Tian Y, Wang YW, Peng X, and Zhang W A fault diagnosis method for few-shot industrial processes based on semantic segmentation and hybrid domain transfer learning Appl Intell 2023 53 23 28268-90
[32]
An YY, Zhang K, Chai Y, Liu Q, and Huang XH Domain adaptation network base on contrastive learning for bearings fault diagnosis under variable working conditions Expert Syst Appl 2023 212 118802
[33]
Wang QB, Xu YB, Yang SK, Chang JT, Zhang JG, Kong XG (2023) A domain adaptation method for bearing fault diagnosis using multiple incomplete source data. J Intell Manuf
[34]
Su ZH, Zhang JY, Tang JX, Wang YX, Xu HB, Zou JX, and Fan SC A novel deep transfer learning method with inter-domain decision discrepancy minimization for intelligent fault diagnosis Knowl-Based Syst 2023 259 110065
[35]
Li CJ, Li SB, Wang H, Gu FS, and Ball AD Attention-based deep meta-transfer learning for few-shot fine-grained fault diagnosis Knowl-Based Syst 2023 264 110345
[36]
Luo SY, Huang XF, Wang YZ, Luo RM, and Zhou Q Transfer learning based on improved stacked autoencoder for bearing fault diagnosis Knowl-Based Syst 2022 256 109846
[37]
Ren C, Jiang B, Lu NY, Simani S, Gao FR (2023) Meta-Learning With Distributional Similarity Preference for Few-Shot Fault Diagnosis Under Varying Working Conditions. IEEE T Cybernetics
[38]
Yang J and Wang XM Meta-learning with deep flow kernel network for few shot cross-domain remaining useful life prediction Reliab Eng Syst Safe 2024 244 109928
[39]
Wu K, Nie YK, Wu J, and Wang YH Prior knowledge-based self-supervised learning for intelligent bearing fault diagnosis with few fault samples Meas Sci Technol 2023 34 10 105104
[40]
Long JY, Zhang RX, Chen YB, Zhao R, Yang Z, Huang YW, Li C (2023) A Customized Meta-Learning Framework for Diagnosing New Faults From Unseen Working Conditions With Few Labeled Data. IEEE-Asme T Mech
[41]
Zhang T, Jiao JY, Lin J, Li H, Hua JD, and He D Uncertainty-based contrastive prototype-matching network towards cross-domain fault diagnosis with small data Knowl-Based Syst 2022 254 109651
[42]
Ma RY, Han T, and Lei WX Cross-domain meta learning fault diagnosis based on multi-scale dilated convolution and adaptive relation module Knowl-Based Syst 2023 261 110175
[43]
Zheng XQ, Yue CY, Wei J, Xue AK, Ge M, and Kong YG Few-shot intelligent fault diagnosis based on an improved meta-relation network Appl Intell 2023 53 24 30080-96
[44]
Ma L, Jiang B, Xiao L, and Lu N Digital twin-assisted enhanced meta-transfer learning for rolling bearing fault diagnosis Mech Syst Signal Pr 2023 200 110490
[45]
Fan C, Zhang Y, Ma H, Ma Z, Yu K, Zhao S, and Zhang X A novel metric-based model with the ability of zero-shot learning for intelligent fault diagnosis Eng Appl Artif Intel 2024 129 107605
[46]
Chen ZY, Li ZR, Wu J, Deng C, and Dai W Deep residual shrinkage relation network for anomaly detection of rotating machines J Manuf Syst 2022 65 579-590
[47]
Pan XR, Ge CJ, Lu R, Song SJ, Chen GF, Huang ZY, Huang G (2022) On the integration of self-attention and convolution. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 815–825
[48]
Cuturi M Sinkhorn distances: Lightspeed computation of optimal transport Adv Neural Inf Process Syst 2013 26 2292-2300
[49]
Yang B, Lei YG, Xu SC, and Lee CG An Optimal Transport-Embedded Similarity Measure for Diagnostic Knowledge Transferability Analytics Across Machines IEEE T Ind Electron 2022 69 7372-7382
[50]
Gousseau W, Antoni J, Girardin F, and Griffaton J Analysis of the Rolling Element Bearing data set of the Center for Intelligent Maintenance Systems of the University of Cincinnati 2016
[51]
Smith WA and Randall RB Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study Mech Syst Signal Pr 2015 64–65 100-131
[52]
He KM, Zhang XY, Ren SQ, Sun J (2016) Ieee, Deep Residual Learning for Image Recognition, CVPR 2016. pp. 770–778
[53]
Chen ZY, Wang C, Wu J, Deng C, Wang YH (2022) Deep convolutional transfer learning-based structural damage detection with domain adaptation. Appl Intell 1–15
[54]
Wang JH and Zhai YP Prototypical Siamese Networks for Few-shot Learning ICEIEC 2020 2020 178-181
[55]
Vinyals O, Blundell C, Lillicrap T, Kavukcuoglu K, Wierstra D (2016) Matching Networks for One Shot Learning. Adv Neural Inf Process Syst 29
[56]
Chen ZY, Wang YH, Wu J, Deng C, and Jiang WX Wide Residual Relation Network-Based Intelligent Fault Diagnosis of Rotating Machines with Small Samples Sensors-Basel 2022 22 11 4161
[57]
Yang JL, Wang CD, and Wei CA A novel Brownian correlation metric prototypical network for rotating machinery fault diagnosis with few and zero shot learners Adv Eng Inform 2022 54 101815
[58]
Chen ZY, Wu K, Wu J, Deng C, and Wang YH Residual shrinkage transformer relation network for intelligent fault detection of industrial robot with zero-fault samples Knowl-Based Syst 2023 268 110452
[59]
Zhang G, Zeng YJ, Zhang TQ (2023) Second-order coupled tristable stochastic resonance and its application in bearing fault detection under different noises. Nonlinear Dynam 1–23.
Recommendations
On undetectable faults and fault diagnosis
The presence of an undetectable fault ui may modify the response of a detectable fault dj to a test set used for fault diagnosis. This may impact the accuracy of fault diagnosis based on the responses of single faults. Many state-of-the-art diagnosis ...
Intelligent fault diagnosis of rotating machinery based on one-dimensional convolutional neural network
Highlights- A one-dimensional convolutional neural network model is proposed to Diagnostic fault with the original vibration signal.
AbstractFault diagnosis of rotating machinery plays a significant role in the reliability and safety of modern industrial systems. The traditional fault diagnosis methods usually need manually extracting the features from raw sensor data ...
Selection of a fault model for fault diagnosis based on unique responses
In this paper, we describe a preprocessing step to fault diagnosis of an observed response obtained from a faulty chip. In this step, a fault model for diagnosing the observed response is selected. This step allows fault diagnosis to be performed based ...
Comments
Information & Contributors
Information
Published In
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Publisher
Kluwer Academic Publishers
United States
Publication History
Published: 25 May 2024
Accepted: 10 May 2024
Author Tags
Qualifiers
- Research-article
Funding Sources
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 13 Jan 2025