Due to strong background noise and complex operating conditions, deep learning-based fault diagnosis encounters excessive data dependency, prolonged training durations, and suboptimal prediction accuracy. In this article, using an improved residual neural network (ResNet) model, an optimized recurrence plot (RP) (OptRP) method is developed to detect early weak faults in bearings. The technique comprises a multiscale RP, a maximum information entropy RP, and a signed RP. First, in order to address the data redundancy and single-scale limitations of RPs, an asymmetric multiscale RP is constructed. Second, using the maximum information entropy as the optimization criterion, an optimal threshold function is designed to overcome the conventional RP’s inability to automatically select the optimal threshold. Finally, a direction vector is defined to resolve the serious trend confusion problem inherent in the conventional RP algorithm during the phase space reconstruction process, and the OptRP image is generated by multiplying the direction vector matrix with the multiscale, maximum information entropy RP. Furthermore, an improved ResNet model, called as improved residual neural network (IResNet), is proposed by incorporating the Inception module. IResNet effectively enhances the multiscale feature extraction capability of the existing ResNet. The effectiveness of the OptRP-IResNet method is validated on a small sample dataset of early weak faults in high-speed train bearings, achieving a 5.61% improvement in a long-time series fault diagnosis experiment. The superiority of the method is further proved by gradient-weighted class activation mapping visualization results.
AbtaneMDahiKMartinezH, et al. Axle bearing fault diagnosis for high-speed trains: a comprehensive review of methodologies, technologies, challenges and emerging trends. Measurement2025; 251: 117098.
2.
WangBLiuYYangS, et al. Characteristics analysis on bearing rotor system of high-speed train with track irregularity. Mech Syst Signal Process2024; 221: 111674.
3.
LiuZZhangL.A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearings. Measurement2020; 149: 107002.
4.
IsmailFForestierGWeberJ, et al. Deep learning for time series classification: a review. Data Min Knowl Disc2019; 33: 917–963.
5.
TorresJHadjoutDSebaaA, et al. Deep learning for time series forecasting: a survey. Big Data2021; 9: 3–21.
6.
LimBZohrenS.Time-series forecasting with deep learning: a survey. Philos Trans R Soc A Math Phys Eng Sci2021; 379: 20200209.
7.
KongXChenZLiuW, et al. Deep learning for time series forecasting: a survey. Int J Mach Learn Cyber2025; 16: 5079–5112.
8.
ArchanaRJeevarajP.Deep learning models for digital image processing: a review. Artif Intell Rev2024; 57: 11.
9.
TianCFeiLZhengW, et al. Deep learning on image denoising: an overview. Neural Netw2020; 131: 251–275.
10.
WangZChenJHoiS.Deep learning for image super-resolution: a survey. IEEE Trans Pattern Anal Mach Intell2021; 43: 3365–3387.
11.
ShenDWuGSukH.Deep learning in medical image analysis. Annu Rev Biomed Eng2017; 19: 221–248.
12.
ZhaoZZhengPXuS, et al. Object detection with deep learning: a review. IEEE Trans Neural Netw Learn Syst2019; 30: 3212–3232.
13.
KhorramAKhalooeiMRezghiM.End-to-end CNN + LSTM deep learning approach for bearing fault diagnosis. Appl Intell2021; 51: 736–751.
14.
ZhangSZhangSWangB, et al. Deep learning algorithms for bearing fault diagnostics: a comprehensive review. IEEE Access2020; 8: 29857–29881.
15.
ShenSLuHSadoughiM, et al. A physics-informed deep learning approach for bearing fault detection. Eng Appl Artif Intell2021; 103: 104295.
16.
TangHTangYSuY, et al. Feature extraction of multi-sensors for early bearing fault diagnosis using deep learning based on minimum unscented Kalman filter. Eng Appl Artif Intell2024; 127: 107138.
17.
ZouFZhangHSangS, et al. An anti-noise one-dimension convolutional neural network learning model applying on bearing fault diagnosis. Measurement2021; 186: 110236.
18.
OzcanIDeveciogluOInceT, et al. Enhanced bearing fault detection using multichannel, multilevel 1D CNN classifier. Electr Eng2022; 104: 435–447.
WangXMaoDLiX.Bearing fault diagnosis based on vibro-acoustic data fusion and 1D-CNN network. Measurement2021; 173: 108518.
21.
LongZZhangXSongD, et al. Motor fault diagnosis using image visual information and bag of words model. IEEE Sens J2021; 21: 21798–21807.
22.
LongYZhouWLuoY.A fault diagnosis method based on one-dimensional data enhancement and convolutional neural network. Measurement2021; 180: 109532.
23.
ChenLCaoJWuK, et al. Application of generalized frequency response functions and improved convolutional neural network to fault diagnosis of heavy-duty industrial robot. Robot Comput Integr Manuf2022; 73: 102228.
24.
ZhaoJYangSLiQ, et al. A new bearing fault diagnosis method based on signal-to-image mapping and convolutional neural network. Measurement2021; 176: 109088.
25.
ZhangYHuaJZhangD, et al. Train bearing fault diagnosis based on time–frequency signal contrastive domain share CNN. IEEE Sens J2024; 24: 33669–33681.
26.
WangZOatesT.Imaging time-series to improve classification and imputation. In: Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, Buenos Aires, Argentina, 31 July 2015, pp. 3939–3945. NJ: AAAI Press.
27.
SunYLiSWangY, et al. Fault diagnosis of rolling bearing based on empirical mode decomposition and improved Manhattan distance in symmetrized dot pattern image. Mech Syst Signal Process2021; 159: 107817.
28.
WangHXuJYanR, et al. A new intelligent bearing fault diagnosis method using SDP representation and SE-CNN. IEEE Trans Instrum Meas2019; 69: 2377–2389.
29.
LiuXXiaLShiJ, et al. A fault diagnosis method of rolling bearing based on improved recurrence plot and convolutional neural network. IEEE Sens J2023; 23: 10767–10775.
30.
ZhouK.Enhanced feature extraction for machinery condition monitoring using recurrence plot and quantification measure. Int J Adv Manuf Technol2022; 123: 3421–3436.
31.
SharmaA.Fault diagnosis of bearings using recurrences and artificial intelligence techniques. J Nondestruct Eval Diagn Progn Eng Syst2022; 5: 031004.
32.
ZhangYHouYOuYangK, et al. Multi-scale signed recurrence plot based time series classification using inception architectural networks. Pattern Recognit2022; 123: 108385.
33.
ZhouYZhuRZhaoH, et al. Influence of noise on wear fault diagnosis based on recurrence plot. Measurement2022; 205: 112158.
34.
BaiRMengZXuQ, et al. Fractional Fourier and time domain recurrence plot fusion combining convolutional neural network for bearing fault diagnosis under variable working conditions. Reliab Eng Syst Saf2023; 232: 109076.
35.
ShaoKHeYHuX, et al. Distribution recurrence plots and measures: effective signal analysis tools for fault diagnosis of wind turbine drivetrain system. Adv Eng Inform2023; 56: 101985.
36.
WangBQiuWHuX, et al. A rolling bearing fault diagnosis technique based on recurrence quantification analysis and Bayesian optimization SVM. Appl Soft Comput2024; 156: 111506.
37.
HeKZhangXRenS, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition CVPR’16, June 27–30, 2016, Las Vegas, NV, USA, 2016, pp. 770–778. Piscataway, NJ: IEEE.
38.
SzegedyCLiuWJiaY, et al. Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition CVPR’15, June 7–12, 2015, Boston, MA, USA, 2015, pp. 1–9. Piscataway, NJ: IEEE.
39.
EckmannJKamphorstSRuelleD.Recurrence plots of dynamical systems. EPL1987; 4: 973.
40.
MagadanLRuiz-CarcelCGrandaJ, et al. Explainable and interpretable bearing fault classification and diagnosis under limited data. Adv Eng Inform2024; 62: 102909.
WallotSLeonardiG.Analyzing multivariate dynamics using cross-recurrence quantification analysis (CRQA), diagonal-cross-recurrence profiles (DCRP), and multidimensional recurrence quantification analysis (MDRQA)–a tutorial in R. Front Psychol2018; 9: 2232.
43.
MarwanNCarmen RomanoMThielM, et al. Recurrence plots for the analysis of complex systems. Phys Rep2007; 438: 237–329.
44.
LiuBYanCLiuY, et al. Multiscale residual antinoise network via interpretable dynamic recalibration mechanism for rolling bearing fault diagnosis with few samples. IEEE Sens J2023; 23: 31425–31439.
45.
YangCQiaoZLiuL, et al. Positive-incentive noise in artificial intelligence-enabled machine fault diagnosis. Struct Health Monit. Epub ahead of print. 2025.
46.
LabibMINahidA-A.OptRPC: a novel and optimized recurrence plot-based system for ECG beat classification. Biomed Signal Process Control2022; 72: 103328.
47.
XueQTianRWangJ, et al. Periodic-state-directional-transition weak signal detection method and its application. J Vib Eng Technol2025; 13: 582.
48.
ChengZ.Extraction and diagnosis of rolling bearing fault signals based on improved wavelet transform. J Meas Eng2023; 11: 420–436.
49.
SelvarajuRCogswellMDasA, et al. Grad-CAM: Visual explanations from deep networks via gradient-based localization. In : Proceedings of the IEEE international conference on computer vision ICCV’17, October 22–29, 2017, Venice, Italy, 2017, pp. 618–626. Piscataway, NJ: IEEE.
