pith. sign in

arxiv: 2606.20323 · v1 · pith:JOWWV4MTnew · submitted 2026-06-18 · 💻 cs.AI

Leveraging systems' non-linearity to tackle the scarcity of data in the design of Intelligent Fault Diagnosis Systems

Giancarlo Santamato , Andrea Mattia Garavagno , Massimiliano Solazzi , Antonio Frisoli This is my paper

Pith reviewed 2026-06-26 17:05 UTC · model grok-4.3

classification 💻 cs.AI
keywords intelligent fault diagnosisdeep transfer learningdata scarcitynon-linear systemsvibration signalsconvolutional neural networksimage visualizationrailway pantograph
0
0 comments X

The pith

A periodic multi-excitation procedure exploits system non-linearities to generate images that pre-trained CNNs can use for fault diagnosis with scarce labeled data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes a method to build intelligent fault diagnosis systems using deep transfer learning when labeled data is scarce. It introduces a periodic multi-excitation level procedure that takes advantage of real-world systems' non-linear behaviors to create images from vibration signals. These images can then be analyzed by pre-trained convolutional neural networks to identify faults. A new data visualization technique and its augmentation are also presented to address data limitations. The approach is validated experimentally on a railway pantograph structure.

Core claim

The central discovery is that by applying periodic excitations at multiple levels to a system, the resulting non-linear vibration responses can be visualized as images containing diagnostic information. Pre-trained CNNs can then classify these images to diagnose faults even when only a small amount of labeled data is available. This is supported by the proposal of a new visualization method and augmentation technique for the generated data.

What carries the argument

The periodic multi-excitation level procedure that leverages intrinsic non-linearities to produce analyzable images from vibration signals.

Load-bearing premise

The generated images from the multi-excitation procedure contain features that pre-trained CNNs can reliably use to classify faults even with very small amounts of labeled data.

What would settle it

Observing whether classification accuracy on the pantograph fault diagnosis task drops significantly when using standard single-excitation images compared to the multi-excitation images under the same limited data conditions.

read the original abstract

Deep Transfer Learning (DTL) allows for the efficient building of Intelligent Fault Diagnosis Systems (IFDS). On the other hand, DTL methods still heavily rely on large amounts of labelled data. Obtaining such an amount of data can be challenging when dealing with machines or structures faults. This document proposes a novel approach to the design of vibration-based IFDS using DTL in condition of strong data scarcity. A periodic multi-excitation level procedure leveraging intrinsic non-linearities of real-world systems is used to produce images that can be conveniently analysed by pre-trained Convolutional Neural Networks (CNNs) to diagnose faults. A new data visualization method and its augmentation technique are proposed in this paper to tackle the typical lack of data encountered during the design of IFDS. Experimental validation on a railway pantograph structure provides effective support for the proposed method.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper proposes a novel approach for vibration-based Intelligent Fault Diagnosis Systems (IFDS) under strong data scarcity using Deep Transfer Learning (DTL). It introduces a periodic multi-excitation level procedure that exploits intrinsic non-linearities of real-world systems to generate images from vibration signals, which are then analyzed by pre-trained CNNs. A new data visualization method and augmentation technique are presented, with experimental validation on a railway pantograph structure claimed to provide effective support for the method.

Significance. If validated with quantitative evidence, the approach could meaningfully advance IFDS design in data-scarce industrial settings by converting scarce vibration data into image representations amenable to transfer learning, potentially reducing reliance on large labeled fault datasets in applications such as railway infrastructure monitoring.

major comments (2)
  1. [Abstract] Abstract: The claim that 'experimental validation on a railway pantograph structure provides effective support' is load-bearing for the central claim yet supplies no quantitative results, error bars, baseline comparisons, accuracy metrics, or details on the vibration-to-image conversion process, preventing assessment of whether the generated images contain extractable diagnostic features for pre-trained CNNs.
  2. [Abstract] The weakest assumption—that images from the multi-excitation procedure reliably encode fault-specific features extractable by CNNs even with very small labeled data—is not supported by any reported performance numbers or ablation studies in the manuscript description, leaving the method's advantage over standard DTL approaches unquantified.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for these comments on the abstract. We agree that the abstract would benefit from greater specificity regarding quantitative results and will revise it accordingly while preserving its concise nature. We address each point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that 'experimental validation on a railway pantograph structure provides effective support' is load-bearing for the central claim yet supplies no quantitative results, error bars, baseline comparisons, accuracy metrics, or details on the vibration-to-image conversion process, preventing assessment of whether the generated images contain extractable diagnostic features for pre-trained CNNs.

    Authors: We agree that the abstract would be improved by incorporating key quantitative results. The manuscript body reports these details, including accuracy metrics, baseline comparisons, and the vibration-to-image conversion process. We will revise the abstract to include a concise summary of the main performance figures and a brief description of the image generation method. revision: yes

  2. Referee: [Abstract] The weakest assumption—that images from the multi-excitation procedure reliably encode fault-specific features extractable by CNNs even with very small labeled data—is not supported by any reported performance numbers or ablation studies in the manuscript description, leaving the method's advantage over standard DTL approaches unquantified.

    Authors: The manuscript contains performance numbers and ablation studies in the experimental section that quantify the encoding of fault-specific features and the advantage under data scarcity. To make this evident from the abstract alone, we will add a short statement summarizing the key quantitative findings and the observed improvement relative to standard DTL baselines. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation rests on external experimental validation

full rationale

The paper's central claim is a data-generation pipeline (periodic multi-excitation leveraging system non-linearities to create images for pre-trained CNNs) whose validity is asserted via concrete experimental results on a pantograph structure. No equations, definitions, or self-citations are supplied that reduce any prediction or uniqueness claim to a fitted parameter or prior author work by construction. The approach is presented as an empirical method tested on real hardware rather than a self-referential derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Full manuscript text unavailable; no free parameters, axioms, or invented entities can be identified from the abstract. The method implicitly assumes that non-linear vibration responses produce image features separable by existing CNNs, but this is not formalized.

pith-pipeline@v0.9.1-grok · 5686 in / 1134 out tokens · 13897 ms · 2026-06-26T17:05:59.564022+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

64 extracted references · 2 canonical work pages · 2 internal anchors

  1. [1]

    Application of empirical mode decomposition and artificial neural network for automatic bearing fault diagnosis based on vibration signals

    Jaouher Ben Ali, Nader Fnaiech, Lotfi Saidi, Brigitte Chebel-Morello, and Farhat Fnaiech. Application of empirical mode decomposition and artificial neural network for automatic bearing fault diagnosis based on vibration signals. Applied Acoustics, 89:16–27, 2015

    2015

  2. [2]

    Rolling bearing fault diagnosis algorithm using over- lapping group sparse-deep complex convolu- tional neural network

    Fengping An and Jianrong Wang. Rolling bearing fault diagnosis algorithm using over- lapping group sparse-deep complex convolu- tional neural network. Nonlinear Dynamics, 108(3):2353–2368, 2022

    2022

  3. [3]

    A review of vibration-based damage detection in civil structures: From traditional methods to machine learning and deep learning applica- tions

    Onur Avci, Osama Abdeljaber, Serkan Kiranyaz, Mohammed Hussein, Moncef Gab- bouj, and Daniel J Inman. A review of vibration-based damage detection in civil structures: From traditional methods to machine learning and deep learning applica- tions. Mechanical systems and signal process- ing, 147:107077, 2021

    2021

  4. [4]

    Modified chain-code computer vision techniques for interrogation of vibra- tion signatures for structural fault detection

    M Biswas, AK Pandey, SA Bluni, and MM Samman. Modified chain-code computer vision techniques for interrogation of vibra- tion signatures for structural fault detection. Journal of sound and vibration , 175(1):89– 104, 1994

    1994

  5. [5]

    Waveform chain code: a more sensitive feature selection in unsupervised structural damage detec- tion

    Shilei Chen and Zhi Chao Ong. Waveform chain code: a more sensitive feature selection in unsupervised structural damage detec- tion. In 2019 1st International Conference on Artificial Intelligence and Data Sciences (AiDAS), pages 116–119. IEEE, 2019

    2019

  6. [6]

    Unsu- pervised damage identification scheme using pca-reduced frequency response function and waveform chain code analysis

    Shilei Chen, Zhi Chao Ong, Wei Haur Lam, Kok-Sing Lim, and Khin Wee Lai. Unsu- pervised damage identification scheme using pca-reduced frequency response function and waveform chain code analysis. International Journal of Structural Stability and Dynamics, 20(08):2050091, 2020

    2020

  7. [7]

    Randaugment: Prac- tical automated data augmentation with a reduced search space

    Ekin D Cubuk, Barret Zoph, Jonathon Shlens, and Quoc V Le. Randaugment: Prac- tical automated data augmentation with a reduced search space. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops , pages 702–703, 2020

    2020

  8. [8]

    Modal testing: theory, prac- tice and application

    David J Ewins. Modal testing: theory, prac- tice and application . John Wiley & Sons, 2009

    2009

  9. [9]

    Ans-net: anti- noise siamese network for bearing fault diag- nosis with a few data

    Qin Fang and Dinghui Wu. Ans-net: anti- noise siamese network for bearing fault diag- nosis with a few data. Nonlinear Dynamics, Springer Nature 2021 LATEX template 10 Leveraging systems’ non-linearity to tackle the scarcity of data in the design of Intelligent Fault Diagnosis Systems 104(3):2497–2514, 2021

    2021

  10. [10]

    A novel deep learning system with data augmentation for machine fault diagnosis from vibration sig- nals

    Qiang Fu and Huawei Wang. A novel deep learning system with data augmentation for machine fault diagnosis from vibration sig- nals. Applied Sciences, 10(17):5765, 2020

    2020

  11. [11]

    A fault diagnosis method of rolling bearing based on complex morlet cwt and cnn

    Dawei Gao, Yongsheng Zhu, Xian Wang, Ke Yan, and Jun Hong. A fault diagnosis method of rolling bearing based on complex morlet cwt and cnn. In 2018 prognostics and system health management conference (PHM-Chongqing), pages 1101–1105. IEEE, 2018

    2018

  12. [12]

    Data augmentation in fault diagnosis based on the wasserstein generative adversarial net- work with gradient penalty

    Xin Gao, Fang Deng, and Xianghu Yue. Data augmentation in fault diagnosis based on the wasserstein generative adversarial net- work with gradient penalty. Neurocomputing, 396:487–494, 2020

    2020

  13. [13]

    Data augmen- tation for intelligent mechanical fault diagno- sis based on local shared multiple-generator gan

    Qingwen Guo, Yibin Li, Yanjun Liu, Shengyao Gao, and Yan Song. Data augmen- tation for intelligent mechanical fault diagno- sis based on local shared multiple-generator gan. IEEE Sensors Journal , 22(10):9598– 9609, 2022

    2022

  14. [14]

    A novel fault diagnosis method for rotating machinery based on a convolutional neural network

    Sheng Guo, Tao Yang, Wei Gao, and Chen Zhang. A novel fault diagnosis method for rotating machinery based on a convolutional neural network. Sensors, 18(5):1429, 2018

    2018

  15. [15]

    A bearing fault diagno- sis method based on wavelet packet transform and convolutional neural network optimized by simulated annealing algorithm

    Feng He and Qing Ye. A bearing fault diagno- sis method based on wavelet packet transform and convolutional neural network optimized by simulated annealing algorithm. Sensors, 22(4):1410, 2022

    2022

  16. [16]

    Review on the new development of vibration-based damage identification for civil engineering structures: 2010–2019

    Rongrong Hou and Yong Xia. Review on the new development of vibration-based damage identification for civil engineering structures: 2010–2019. Journal of Sound and Vibration , 491:115741, 2021

    2010

  17. [17]

    Data simulation by resampling—a practical data augmentation algorithm for periodical signal analysis-based fault diagnosis

    Tianhao Hu, Tang Tang, and Ming Chen. Data simulation by resampling—a practical data augmentation algorithm for periodical signal analysis-based fault diagnosis. Ieee Access, 7:125133–125145, 2019

    2019

  18. [18]

    A simple data augmentation algorithm and a self-adaptive convolutional architecture for few-shot fault diagnosis under different working conditions

    Tianhao Hu, Tang Tang, Ronglai Lin, Ming Chen, Shufa Han, and Jie Wu. A simple data augmentation algorithm and a self-adaptive convolutional architecture for few-shot fault diagnosis under different working conditions. Measurement, 156:107539, 2020

    2020

  19. [19]

    Automated bearing fault diagnosis scheme using 2d representation of wavelet packet transform and deep convolutional neural net- work

    MM Manjurul Islam and Jong-Myon Kim. Automated bearing fault diagnosis scheme using 2d representation of wavelet packet transform and deep convolutional neural net- work. Computers in Industry , 106:142–153, 2019

    2019

  20. [20]

    Adam: A Method for Stochastic Optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  21. [21]

    Vibration based health monitoring for a lightweight truss structure: experimental assessment of several statistical time series methods

    Fotis P Kopsaftopoulos and Spilios D Fas- sois. Vibration based health monitoring for a lightweight truss structure: experimental assessment of several statistical time series methods. Mechanical Systems and Signal Processing, 24(7):1977–1997, 2010

    1977

  22. [22]

    Permutation entropy-based 2d feature extraction for bearing fault diag- nosis

    Mantas Landauskas, Maosen Cao, and Min- vydas Ragulskis. Permutation entropy-based 2d feature extraction for bearing fault diag- nosis. Nonlinear Dynamics , 102:1717–1731, 2020

    2020

  23. [23]

    A metric on the similarity between two frequency response functions

    Dooho Lee, Tae-Soo Ahn, and Hyeon-Seok Kim. A metric on the similarity between two frequency response functions. Journal of Sound and Vibration , 436:32–45, 2018

    2018

  24. [24]

    Multi-mode data augmentation and fault diagnosis of rotat- ing machinery using modified acgan designed with new framework

    Wei Li, Xiang Zhong, Haidong Shao, Baop- ing Cai, and Xingkai Yang. Multi-mode data augmentation and fault diagnosis of rotat- ing machinery using modified acgan designed with new framework. Advanced Engineering Informatics, 52:101552, 2022

    2022

  25. [25]

    A perspective survey on deep transfer learning for fault diagnosis in industrial scenarios: Theories, applications and challenges

    Weihua Li, Ruyi Huang, Jipu Li, Yixiao Liao, Zhuyun Chen, Guolin He, Ruqiang Yan, and Konstantinos Gryllias. A perspective survey on deep transfer learning for fault diagnosis in industrial scenarios: Theories, applications and challenges. Mechanical Systems and Signal Processing, 167:108487, 2022. Springer Nature 2021 LATEX template Leveraging systems’ n...

    2022

  26. [26]

    Gear pitting fault diag- nosis with mixed operating conditions based on adaptive 1d separable convolution with residual connection

    Xueyi Li, Jialin Li, Chengying Zhao, Yongzhi Qu, and David He. Gear pitting fault diag- nosis with mixed operating conditions based on adaptive 1d separable convolution with residual connection. Mechanical Systems and Signal Processing, 142:106740, 2020

    2020

  27. [27]

    A fault diagnosis scheme for rolling bearing based on local mean decompo- sition and improved multiscale fuzzy entropy

    Yongbo Li, Minqiang Xu, Rixin Wang, and Wenhu Huang. A fault diagnosis scheme for rolling bearing based on local mean decompo- sition and improved multiscale fuzzy entropy. Journal of Sound and Vibration , 360:277– 299, 2016

    2016

  28. [28]

    The interpolation damage detection method for frames under seismic excitation

    Maria Pina Limongelli. The interpolation damage detection method for frames under seismic excitation. Journal of Sound and Vibration, 330(22):5474–5489, 2011

    2011

  29. [29]

    Frequency response func- tion interpolation for damage detection under changing environment

    MP Limongelli. Frequency response func- tion interpolation for damage detection under changing environment. Mechanical Sys- tems and Signal Processing, 24(8):2898–2913, 2010

    2010

  30. [30]

    Lightweight transfer nets and adversarial data augmentation for photovoltaic series arc fault detection with limited fault data

    Shibo Lu, Rui Ma, Tharmakulasingam Sirojan, BT Phung, and Daming Zhang. Lightweight transfer nets and adversarial data augmentation for photovoltaic series arc fault detection with limited fault data. International Journal of Electrical Power & Energy Systems, 130:107035, 2021

    2021

  31. [31]

    Nonlinear system identification in structural dynamics: 10 more years of progress.Mechan- ical Systems and Signal Processing , 83:2–35, 2017

    Jean-Philippe No¨ el and Ga¨ etan Kerschen. Nonlinear system identification in structural dynamics: 10 more years of progress.Mechan- ical Systems and Signal Processing , 83:2–35, 2017

    2017

  32. [32]

    Convolutional neural network based rolling-element bearing fault diagnosis for naturally occurring and progressing defects using time-frequency domain features

    Vibhor Pandhare, Jaskaran Singh, and Jay Lee. Convolutional neural network based rolling-element bearing fault diagnosis for naturally occurring and progressing defects using time-frequency domain features. In 2019 Prognostics and System Health Manage- ment Conference (PHM-Paris) , pages 320–

    2019

  33. [33]

    Effectiveness of the frf curvature technique for structural health monitoring

    MC Porcu, DM Patteri, S Melis, and F Aymerich. Effectiveness of the frf curvature technique for structural health monitoring. Construction and Building Materials , 226: 173–187, 2019

    2019

  34. [34]

    Bearing fault diagnosis based on convolutional neural networks with kurtogram representation of acoustic emis- sion signals

    Alexander Prosvirin, JaeYoung Kim, and Jong-Myon Kim. Bearing fault diagnosis based on convolutional neural networks with kurtogram representation of acoustic emis- sion signals. In Advances in Computer Sci- ence and Ubiquitous Computing: CSA-CUTE 17, pages 21–26. Springer, 2018

    2018

  35. [35]

    Ima- genet large scale visual recognition challenge

    Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, et al. Ima- genet large scale visual recognition challenge. International journal of computer vision, 115: 211–252, 2015

    2015

  36. [36]

    Mobilenetv2: Inverted residuals and linear bottlenecks

    Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4510–4520, 2018

    2018

  37. [37]

    Detecting early damages in the rail- way pantograph mechanism: a multiple exci- tation approach for the frequency domain

    G Santamato, D Chiaradia, M Solazzi, and A Frisoli. Detecting early damages in the rail- way pantograph mechanism: a multiple exci- tation approach for the frequency domain. Vehicle System Dynamics, pages 1–22, 2023

    2023

  38. [38]

    A lightweight robotic device for the inspec- tion of railway pantograph

    Giancarlo Santamato, Domenico Chiaradia, Massimiliano Solazzi, and Antonio Frisoli. A lightweight robotic device for the inspec- tion of railway pantograph. In 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pages 284–289. IEEE, 2019

    2019

  39. [39]

    A lightweight robotic device based on a micro- macro actuation concept for the inspection of railway pantograph

    Giancarlo Santamato, Domenico Chiaradia, Massimiliano Solazzi, and Antonio Frisoli. A lightweight robotic device based on a micro- macro actuation concept for the inspection of railway pantograph. Journal of Mechanisms and Robotics, 12(6):061002, 2020

    2020

  40. [40]

    Investigating the effect of dry-friction on damage detection tests

    Giancarlo Santamato, Massimiliano Solazzi, and Antonio Frisoli. Investigating the effect of dry-friction on damage detection tests. Journal of Sound and Vibration , 568:117949, 2024. Springer Nature 2021 LATEX template 12 Leveraging systems’ non-linearity to tackle the scarcity of data in the design of Intelligent Fault Diagnosis Systems

    2024

  41. [41]

    Parametric and nonparametric identification of linear systems in the presence of nonlinear distortions-a frequency domain approach

    Johan Schoukens, Tadeusz Dobrowiecki, and Rik Pintelon. Parametric and nonparametric identification of linear systems in the presence of nonlinear distortions-a frequency domain approach. IEEE Transactions on Automatic Control, 43(2):176–190, 1998

    1998

  42. [42]

    A novel method for intelli- gent fault diagnosis of rolling bearings using ensemble deep auto-encoders

    Haidong Shao, Hongkai Jiang, Ying Lin, and Xingqiu Li. A novel method for intelli- gent fault diagnosis of rolling bearings using ensemble deep auto-encoders. Mechanical Systems and Signal Processing , 102:278–297, 2018

    2018

  43. [43]

    Rolling bearing fault fea- ture learning using improved convolutional deep belief network with compressed sensing

    Haidong Shao, Hongkai Jiang, Haizhou Zhang, Wenjing Duan, Tianchen Liang, and Shuaipeng Wu. Rolling bearing fault fea- ture learning using improved convolutional deep belief network with compressed sensing. Mechanical Systems and Signal Processing , 100:743–765, 2018

    2018

  44. [44]

    Gen- erative adversarial networks for data augmen- tation in machine fault diagnosis

    Siyu Shao, Pu Wang, and Ruqiang Yan. Gen- erative adversarial networks for data augmen- tation in machine fault diagnosis. Computers in Industry, 106:85–93, 2019

    2019

  45. [45]

    Bearing fault diagnosis based on optimal convolution neu- ral network

    Yongjian Sun and Shaohui Li. Bearing fault diagnosis based on optimal convolution neu- ral network. Measurement, 190:110702, 2022

    2022

  46. [46]

    Distance and density similarity based enhanced-nn classi- fier for improving fault diagnosis performance of bearings

    Sharif Uddin, Md Islam, Sheraz Ali Khan, Jaeyoung Kim, Jong-Myon Kim, Seok-Man Sohn, Byeong-Keun Choi, et al. Distance and density similarity based enhanced-nn classi- fier for improving fault diagnosis performance of bearings. Shock and Vibration, 2016, 2016

    2016

  47. [47]

    Deep learning enabled fault diagnosis using time-frequency image analysis of rolling element bearings

    David Verstraete, Andr´ es Ferrada, Enrique L´ opez Droguett, Viviana Meruane, Mohammad Modarres, et al. Deep learning enabled fault diagnosis using time-frequency image analysis of rolling element bearings. Shock and Vibration, 2017, 2017

    2017

  48. [48]

    A supervised sparsity-based wavelet feature for bearing fault diagnosis

    Cong Wang, Meng Gan, and Chang’an Zhu. A supervised sparsity-based wavelet feature for bearing fault diagnosis. Journal of Intel- ligent Manufacturing, 30:229–239, 2019

    2019

  49. [49]

    Intelligent fault diagnosis of planetary gearbox based on refined composite hierarchical fuzzy entropy and random forest

    Yu Wei, Yuantao Yang, Minqiang Xu, and Wenhu Huang. Intelligent fault diagnosis of planetary gearbox based on refined composite hierarchical fuzzy entropy and random forest. ISA transactions, 109:340–351, 2021

    2021

  50. [50]

    The use of fast fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms

    Peter Welch. The use of fast fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms. IEEE Transactions on audio and electroacoustics, 15(2):70–73, 1967

    1967

  51. [51]

    A new snapshot ensemble convolutional neural network for fault diagnosis

    Long Wen, Liang Gao, and Xinyu Li. A new snapshot ensemble convolutional neural network for fault diagnosis. Ieee Access, 7: 32037–32047, 2019

    2019

  52. [52]

    Nonlinearity in structural dynamics: detection, identification and mod- eling

    Keith Worden, Geoffrey R Tomlinson, and K Yagasaki. Nonlinearity in structural dynamics: detection, identification and mod- eling. Appl. Mech. Rev. , 55(2):B26–B27, 2002

    2002

  53. [53]

    Fault diagno- sis for rotating machinery using multiple sensors and convolutional neural networks

    Min Xia, Teng Li, Lin Xu, Lizhi Liu, and Clarence W De Silva. Fault diagno- sis for rotating machinery using multiple sensors and convolutional neural networks. IEEE/ASME transactions on mechatronics , 23(1):101–110, 2017

    2017

  54. [54]

    Cutmix: Regularization strategy to train strong classifiers with local- izable features

    Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo. Cutmix: Regularization strategy to train strong classifiers with local- izable features. In Proceedings of the IEEE/CVF international conference on com- puter vision, pages 6023–6032, 2019

    2019

  55. [55]

    mixup: Beyond Empirical Risk Minimization

    Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  56. [56]

    A new bearing fault diagnosis method based on modified convolutional neural networks

    Jiangquan Zhang, Sun Yi, GUO Liang, GAO Hongli, HONG Xin, and SONG Hongliang. A new bearing fault diagnosis method based on modified convolutional neural networks. Chi- nese Journal of Aeronautics , 33(2):439–447, 2020

    2020

  57. [57]

    An intelligent fault diagnosis method of rolling bearings based on short-time fourier transform and convolutional neural network

    Qi Zhang and Linfeng Deng. An intelligent fault diagnosis method of rolling bearings based on short-time fourier transform and convolutional neural network. Journal of Springer Nature 2021 LATEX template Leveraging systems’ non-linearity to tackle the scarcity of data in the design of Intelligent Fault Diagnosis Systems 13 Failure Analysis and Prevention,...

    2021

  58. [58]

    A deep convolutional neural network with new training methods for bearing fault diagno- sis under noisy environment and different working load

    Wei Zhang, Chuanhao Li, Gaoliang Peng, Yuanhang Chen, and Zhujun Zhang. A deep convolutional neural network with new training methods for bearing fault diagno- sis under noisy environment and different working load. Mechanical systems and signal processing, 100:439–453, 2018

    2018

  59. [59]

    Fault diagno- sis of rotating machinery based on recurrent neural networks

    Yahui Zhang, Taotao Zhou, Xufeng Huang, Longchao Cao, and Qi Zhou. Fault diagno- sis of rotating machinery based on recurrent neural networks. Measurement, 171:108774, 2021

    2021

  60. [60]

    A weighted multi-scale dictionary learning model and its applications on bearing fault diagnosis

    Zhibin Zhao, Baijie Qiao, Shibin Wang, Zhix- ian Shen, and Xuefeng Chen. A weighted multi-scale dictionary learning model and its applications on bearing fault diagnosis. Jour- nal of Sound and Vibration , 446:429–452, 2019

    2019

  61. [61]

    Composite multi-scale phase reverse permutation entropy and its appli- cation to fault diagnosis of rolling bearing

    Jinde Zheng, Yan Chen, Haiyang Pan, and Jinyu Tong. Composite multi-scale phase reverse permutation entropy and its appli- cation to fault diagnosis of rolling bearing. Nonlinear Dynamics, 111(1):459–479, 2023

    2023

  62. [62]

    Random erasing data augmentation

    Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang. Random erasing data augmentation. In Proceedings of the AAAI conference on artificial intelligence , pages 13001–13008, 2020

    2020

  63. [63]

    Deep convolutional generative adversar- ial network with semi-supervised learning enabled physics elucidation for extended gear fault diagnosis under data limitations

    Kai Zhou, Edward Diehl, and Jiong Tang. Deep convolutional generative adversar- ial network with semi-supervised learning enabled physics elucidation for extended gear fault diagnosis under data limitations. Mechanical Systems and Signal Processing , 185:109772, 2023

    2023

  64. [64]

    A multi-layer spiking neural network-based approach to bearing fault diagnosis

    Lin Zuo, Fengjie Xu, Changhua Zhang, Tang- fan Xiahou, and Yu Liu. A multi-layer spiking neural network-based approach to bearing fault diagnosis. Reliability Engineering & System Safety, 225:108561, 2022

    2022