Abstract
The fault diagnosis of gear is indeed a crucial aspect of maintaining rotating machinery, as it helps in ensuring the safe and efficient operation of industrial equipment. Deep learning models have gained significant attention for gear fault diagnosis due to their ability to automatically extract features from raw data, but they also come with their own set of challenges. One major limitation of existing methods is the insufficient consideration given to the impact of environmental noise at industrial field on the diagnostic effectiveness of the models. Additionally, there is a contradiction between the week computational resources of current embedded platforms for industrial field device applications and the large number of parameters and computations required for deep learning models. This may hinder the deployment of complex models in industrial field devices. To address these issues, a novel approach to multi-path convolutional neural network with dual branch attention (AMPCNN) has been proposed. This approach aims to enhance the recognition of different fault types and maintain high accuracy in noisy environments by extracting multi-scale features of the original vibration signal using multi-path convolution and dual branch attention mechanisms. Furthermore, a multi-knowledge distillation (MKD) method has been introduced to construct lightweight multi-sensor gear fault diagnosis models. This approach facilitates the transfer of multiple knowledge from a complex teacher network to a simpler student network, resulting in a lightweight model that exhibits excellent robustness in various noise environments. The experimental results show that the lightweight model achieves high accuracy while requiring significantly fewer floating-point operations and parameter quantities compared to the original teacher network.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Data Availability
The data is available
Materials Availability
The materials is available
Code Availability
The code is available
References
Peng B, Xia H, Lv X et al (2022) An intelligent fault diagnosis method for rotating machinery based on data fusion and deep residual neural network. Appl Intell 52:3051–3065. https://doi.org/10.1007/s10489-021-02555-4
Wang Z, Zhou J, Du W, Lei Y, Wang J (2022) Bearing fault diagnosis method based on adaptive maximum cyclostationarity blind deconvolution. Mech Syst Signal Process 162:108018. https://doi.org/10.1016/j.ymssp.2021.108018
Zhang K, Wang J, Shi H, Zhang X, Tang Y (2021) A fault diagnosis method based on improved convolutional neural network for bearings under variable working conditions. Measurement 182:109749. https://doi.org/10.1016/j.measurement.2021.109749
Karabacak YE, Gürsel Özmen N, Gümüşel L (2022) Intelligent worm gearbox fault diagnosis under various working conditions using vibration, sound and thermal features. Appl Acoust 186:108463. https://doi.org/10.1016/j.apacoust.2021.108463
Saufi SR, Ahmad ZAB, Leong MS, Lim MH (2020) Gearbox fault diagnosis using a deep learning model with limited data sample. IEEE Trans Ind Inform 16:6263–6271. https://doi.org/10.1109/TII.2020.2967822
Xiao D, Ding J, Li X, Huang L (2019) Gear fault diagnosis based on kurtosis criterion vmd and som neural network. Appl Sci 9:5424. https://doi.org/10.3390/app9245424
Tama BA, Vania M, Lee S, Lim S (2023) Recent advances in the application of deep learning for fault diagnosis of rotating machinery using vibration signals. Artif Intell Rev 56:4667–4709. https://doi.org/10.1007/s10462-022-10293-3
Diao N, Wang Z, Ma H, Yang W (2023) Fault diagnosis of rolling bearing under variable working conditions based on cwt and t-resnet. J Vib Eng Technol 11:3747–3757. https://doi.org/10.1007/s42417-022-00780-w
Amarouayache IIE, Saadi MN, Guersi N, Boutasseta N (2020) Bearing fault diagnostics using eemd processing and convolutional neural network methods. Int J Adv Manuf Technol 107:4077–4095. https://doi.org/10.1007/s00170-020-05315-9
Yu X, Ren X, Wan H, Wu S, Ding E (2019) Rolling bearing fault feature extraction and diagnosis method based on modwpt and dbn. In: 2019 11th International conference on Wireless Communications and Signal Processing (WCSP). IEEE, Xi’an, China pp 1–7
Chen B, Shen B, Chen F, Tian H, Xiao W, Zhang F, Zhao C (2019) Fault diagnosis method based on integration of rssd and wavelet transform to rolling bearing. Measurement 131:400–411. https://doi.org/10.1016/j.measurement.2018.07.043
Qin Y, Wang X, Zou J (2019) The optimized deep belief networks with improved logistic sigmoid units and their application in fault diagnosis for planetary gearboxes of wind turbines. IEEE Trans Ind Electron 66:3814–3824. https://doi.org/10.1109/TIE.2018.2856205
Pang S, Yang X (2019) A cross-domain stacked denoising autoencoders for rotating machinery fault diagnosis under different working conditions. IEEE Access 7:77277–77292. https://doi.org/10.1109/ACCESS.2019.2919535
Aljemely AH, Xuan J, Xu L et al (2021) Wise-local response convolutional neural network based on naïve bayes theorem for rotating machinery fault classification. Appl Intell 51:6932–6950. https://doi.org/10.1007/s10489-021-02252-2
Wang C, Li H, Zhang K, Hu S, Sun B (2021) Intelligent fault diagnosis of planetary gearbox based on adaptive normalized cnn under complex variable working conditions and data imbalance. Measurement 180:109565. https://doi.org/10.1016/j.measurement.2021.109565
Hu P, Zhao C, Huang J, Song T (2023) Intelligent and small samples gear fault detection based on wavelet analysis and improved cnn. Processes 11:2969. https://doi.org/10.3390/pr11102969
Zhang Y, Liu W, Wang X, Gu H (2022) A novel wind turbine fault diagnosis method based on compressed sensing and dtl-cnn. Renew Energy 194:249–258. https://doi.org/10.1016/j.renene.2022.05.085
Li T, Zhao Z, Sun C, Yan R, Chen X (2020) Multi-scale cnn for multi-sensor feature fusion in helical gear fault detection. Procedia Manuf 49:89–93. https://doi.org/10.1016/j.promfg.2020.07.001
Wang T, Xu X, Pan H, Chang X, Yuan T, Zhang X, Xu H (2022) Rolling bearing fault diagnosis based on depth-wise separable convolutions with multi-sensor data weighted fusion. Appl Sci 12:7640. https://doi.org/10.3390/app12157640
Li G, Wu J, Deng C, Chen Z (2022) Parallel multi-fusion convolutional neural networks based fault diagnosis of rotating machinery under noisy environments. ISA Trans 128:545–555. https://doi.org/10.1016/j.isatra.2021.10.023
Nguyen CD, Ahmad Z, Kim J-M (2021) Gearbox fault identification framework based on novel localized adaptive denoising technique, wavelet-based vibration imaging, and deep convolutional neural network. Appl Sci 11:7575. https://doi.org/10.3390/app11167575
Zhong X, Li Y, Xia T (2023) Parallel learning attention-guided cnn for signal denoising and mechanical fault diagnosis. J Braz Soc Mech Sci Eng 45:239. https://doi.org/10.1007/s40430-023-04139-4
Liang P, Deng C, Wu J, Yang Z (2020) Intelligent fault diagnosis of rotating machinery via wavelet transform, generative adversarial nets and convolutional neural network. Measurement 159:107768. https://doi.org/10.1016/j.measurement.2020.107768
Ji M, Peng G, Li S et al (2022) A neural network compression method based on knowledge-distillation and parameter quantization for the bearing fault diagnosis. Appl Soft Comput 127:109331. https://doi.org/10.1016/j.asoc.2022.109331
Cheng Y, Wang D, Zhou P, Zhang T (2020) A survey of model compression and acceleration for deep neural networks. https://arxiv.org/abs/1710.09282
Dong Z, Duan Y, Zhou Y et al (2024) Weight-adaptive channel pruning for cnns based on closeness-centrality modeling. Appl Intell 54:201–215. https://doi.org/10.1007/s10489-023-05164-5
Kryzhanovskiy V, Balitskiy G, Kozyrskiy N, Zuruev A (2021) Qpp: real-time quantization parameter prediction for deep neural networks. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Nashville, TN, USA, pp 10679–10687
Li Z, Yang Y (2023) Structurally incoherent adaptive weighted low-rank matrix decomposition for image classification. Appl Intell 53:25028–25041. https://doi.org/10.1007/s10489-023-04875-z
Hinton G, Vinyals O, Dean J (2015) Distilling the knowledge in a neural network. https://arxiv.org/abs/1503.02531
Gou J, Yu B, Maybank SJ, Tao D (2021) Knowledge distillation: a survey. Int J Comput Vis 129:1789–1819. https://doi.org/10.1007/s11263-021-01453-z
Zhang X, He C, Lu Y et al (2022) Fault diagnosis for small samples based on attention mechanism. Measurement 187:110242. https://doi.org/10.1016/j.measurement.2021.110242
Liu W, Rabinovich A, Berg AC (2015) ParseNet: looking wider to see better. https://arxiv.org/abs/1506.04579
Ma N, Zhang X, Liu M, Sun J (2021) Activate or not: learning customized activation. https://arxiv.org/abs/2009.04759
Wang L, Shao Y (2020) Fault feature extraction of rotating machinery using a reweighted complete ensemble empirical mode decomposition with adaptive noise and demodulation analysis. Mech Syst Signal Process 138:106545. https://doi.org/10.1016/j.ymssp.2019.106545
Wang M, Yang Y, Wei L, Li Y (2024) A lightweight gear fault diagnosis method based on attention mechanism and multilayer fusion network. IEEE Trans Instrum Meas 73:1–11. https://doi.org/10.1109/TIM.2023.3330231
Liu H, Chen J, Zhu J, Luo S, Fan H, Xi X (2023) Bearing fault diagnosis and localization method based on wdcnn and unity3d. In: 2023 Panda Forum on Power and Energy (PandaFPE). IEEE, Chengdu, China, pp 359–364
Liang P, Wang W, Yuan X, Liu S, Zhang L, Cheng Y (2022) Intelligent fault diagnosis of rolling bearing based on wavelet transform and improved resnet under noisy labels and environment. Eng Appl Artif Intell 115:105269. https://doi.org/10.1016/j.engappai.2022.105269
Li J, Liu Y, Li Q (2022) Intelligent fault diagnosis of rolling bearings under imbalanced data conditions using attention-based deep learning method. Measurement 189:110500. https://doi.org/10.1016/j.measurement.2021.110500
Ren J, Cai C, Chi Y, Xue Y (2022) Integrated damage location diagnosis of frame structure based on convolutional neural network with inception module. Sensors 23:418. https://doi.org/10.3390/s23010418
Yu W, Lv P (2021) An end-to-end intelligent fault diagnosis application for rolling bearing based on mobilenet. IEEE Access 9:41925–41933. https://doi.org/10.1109/ACCESS.2021.3065195
Jian T, Cao J, Liu W, Xu G, Zhong J (2025) A novel wind turbine fault diagnosis method based on compressive sensing and lightweight squeezenet model. Expert Syst Appl 260:125440. https://doi.org/10.1016/j.eswa.2024.125440
Funding
Not applicable
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. The Implementations, Material preparation, data collection were performed by T.M. Chen, Y.L. Jiang, J.C. Yao, and M. Li. The first draft of the manuscript was written by T.M. Chen and thanks to M.Y. Wang for his revision of the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Some journals require declarations to be submitted in a standardised format. Please check the Instructions for Authors of the journal to which you are submitting to see if you need to complete this section. If yes, your manuscript must contain the following sections under the heading ‘Declarations’:
Conflict of Interest
The authors declare no conflict of interest
Ethics Approval and Consent to participate
This article does not contain studies with human participants or animals. So not applicable
Consent for publication
Informed consent of all participants
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
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.
About this article
Cite this article
Chen, T., Wang, M., Jiang, Y. et al. A lightweight diagnosis method for gear fault based on multi-path convolutional neural networks with attention mechanism. Appl Intell 55, 114 (2025). https://doi.org/10.1007/s10489-024-06094-6
Accepted:
Published:
DOI: https://doi.org/10.1007/s10489-024-06094-6