Restricted accessResearch articleFirst published online 2026
A novel lightweight classification model based on dynamic separation representation and adaptive mapping decision for mechanical bearing fault diagnosis under strong noise interference
Mechanical bearing fault diagnosis is crucial for ensuring the reliability and safety of industrial equipment. However, in the presence of strong noise interference, existing methods face challenges such as high resource consumption, limited generalization across different operating modes of mechanical bearings (time-varying or constant speed), and insufficient capabilities in feature learning and decision-making. To address these issues, this paper proposes a lightweight classification model based on dynamic separation representation and adaptive mapping decision for mechanical bearing fault diagnosis under strong noise interference (LCMFMB). The core components include dynamic asymmetric calibration convolution (DACConv) and adaptive decision classification block (ADCB). First, DACConv effectively identifies and extracts the intrinsic feature information most relevant to bearing health from noisy signals under different bearing operating modes, improving the precision and diversity of feature learning. Then, the feature extraction layer, centered on DACConv, further refines fault features with high discriminability while reducing computational overhead and progressively represents fault information in a stacked manner to form high-dimensional features. Finally, during the classification decision stage, ADCB’s adaptive coordination ability reveals the complex underlying relationships between high-dimensional features and fault types, achieving high-accuracy fault diagnosis. We conducted comparative experiments on publicly available bearing fault diagnosis datasets under variable and constant-speed conditions (compared with seven recent methods), and the competitive experimental results fully validate the effectiveness, generalization, and lightweight nature of LCMFMB under noise interference. Ablation experiments also demonstrate the effectiveness of its core components.
LiYMenZBaiX, et al. A bearing fault diagnosis method based on M-SSCNN and M-LR attention mechanism. Struct Health Monit2025; 24(2): 830–852.
2.
LiXYuanPYuJ, et al. Integrating the maximum mean discrepancy metric with time–frequency enhanced convolutional neural networks for fault diagnosis. Struct Health Monit2024; 0(0).
3.
ShiYWangFHuangY.An adaptive cyclostationarity feature mode decomposition for rolling bearing fault diagnosis under strong background noise. Measurement2025; 256: 118239.
4.
LiSJiJCXuY, et al. IFD-MDCN: multibranch denoising convolutional networks with improved flow direction strategy for intelligent fault diagnosis of rolling bearings under noisy conditions. Reliab Eng Syst Saf2023; 237: 109387.
5.
ZhuangJYanJHuangCG, et al. Residual attention temporal recurrent network for fault diagnosis of gearboxes under limited labeled data. Eng Appl Artif Intell2024; 129: 107539.
6.
LiuJXieFZhangQ, et al. A multisensory time-frequency features fusion method for rotating machinery fault diagnosis under nonstationary case. J Intell Manuf2024; 35(7): 3197–3217.
7.
GuoBQiaoZZhangN, et al. Attention-based ConvNeXt with a parallel multiscale dilated convolution residual module for fault diagnosis of rotating machinery. Expert Syst Appl2024; 249: 123764.
8.
CuiXZhanHHanK, et al. Multi-bearing fault diagnosis method based on convolutional autoencoder causal decoupling domain generalization. ISA Trans2025; 163: 236–250.
9.
JiaoJZhaoMLinJ, et al. A comprehensive review on convolutional neural network in machine fault diagnosis. Neurocomputing2020; 417: 36–63.
10.
FanCZhangYMaH, et al. A novel deep denoising model integrating transformer and time–frequency loss for gearbox fault diagnosis. Adv Eng Inform2025; 66: 103400.
11.
RuanDWangJYanJ, et al. CNN parameter design based on fault signal analysis and its application in bearing fault diagnosis. Adv Eng Inform2023; 55: 101877.
12.
XiaoYShaoHLinJ, et al. BCE-FL: a secure and privacy-preserving federated learning system for device fault diagnosis under non-IID condition in IIoT. IEEE Internet Things J2023; 11(8): 14241–14252.
13.
ZhangYRenZZhouS, et al. Supervised contrastive learning-based domain adaptation network for intelligent unsupervised fault diagnosis of rolling bearing. IEEE/ASME Trans Mechatronics2022; 27(6): 5371–5380.
14.
ChenQDongXTuG, et al. TFN: an interpretable neural network with time-frequency transform embedded for intelligent fault diagnosis. Mech Syst Signal Process2024; 207: 110952.
15.
ZhengXYangPYanK, et al. Rolling bearing fault diagnosis based on multiple wavelet coefficient dimensionality reduction and improved residual network. Eng Appl Artif Intell2024; 133: 108087.
16.
DongXLiGJiaY, et al. Non-iterative denoising algorithm for mechanical vibration signal using spectral graph wavelet transform and detrended fluctuation analysis. Mech Syst Signal Process2021; 149: 107202.
17.
YangBLiuRChenX.Fault diagnosis for a wind turbine generator bearing via sparse representation and shift-invariant K-SVD. IEEE Trans Ind Inform2017; 13(3): 1321–1331.
18.
HuQSiXSZhangQH, et al. A rotating machinery fault diagnosis method based on multi-scale dimensionless indicators and random forests. Mech Syst Signal Process2020; 139: 106609.
19.
WangHLiuZPengD, et al. Interpretable convolutional neural network with multilayer wavelet for Noise-Robust Machinery fault diagnosis. Mech Syst Signal Process2023; 195: 110314.
20.
ZhaoZJiaoY.A fault diagnosis method for rotating machinery based on CNN with mixed information. IEEE Trans Ind Inform2022; 19(8): 9091–9101.
21.
HanBWangHLiuZ, et al. Intelligent vibration signal denoising method based on non-local fully convolutional neural network for rolling bearings. ISA Trans2021; 122: 13–23.
22.
JiaLChowTWSYuanY.GTFE-Net: a Gramian time frequency enhancement CNN for bearing fault diagnosis. Eng Appl Artif Intell2023; 119: 105794.
23.
LiangPWangWYuanX, et al. Intelligent fault diagnosis of rolling bearing based on wavelet transform and improved ResNet under noisy labels and environment. Eng Appl Artif Intell2022; 115: 105269.
24.
HuoTDengLZhangB, et al. Novel imbalanced subdomain adaption multiscale convolutional network for cross-domain unsupervised fault diagnosis of rolling bearings. Meas Sci Technol2023; 35(1): 015905.
25.
LiuSChenJHeS, et al. Few-shot learning under domain shift: Attentional contrastive calibrated transformer of time series for fault diagnosis under sharp speed variation. Mechanical systems and signal processing 2023; 189: 110071.
26.
QinYZhaoLChenY, et al. Learnable wavelet-driven physically interpretable networks for bearing fault diagnosis under variable speed. Mech Syst Signal Process2025; 237: 113121.
27.
WuQDingXGaoS, et al. An order sparse filtering network for interpretable smart gear edge diagnosis under varying speed conditions. Adv Eng Inform2025; 67: 103516.
28.
ZhangXZhangXLiangW, et al. Research on rolling bearing fault diagnosis based on parallel depthwise separable ResNet neural network with attention mechanism. Expert Syst Appl2025; 286: 128105.
29.
YeMYanXJiangD, et al. MIFDELN: a multi-sensor information fusion deep ensemble learning network for diagnosing bearing faults in noisy scenarios. Knowl Based Syst2024; 284: 111294.
30.
LeeSKimT.FRFconv-TDSNet: lightweight, noise-robust convolutional neural network leveraging full-receptive-field convolution and time domain statistics for intelligent machine fault diagnosis. IEEE Trans Instrum Meas2024; 73: 1–13.
31.
LiRWuJLiY, et al. PeriodNet: Noise-robust fault diagnosis method under varying speed conditions. IEEE Trans Neural Netw Learn Syst2023; 35(10): 14045–14059.
32.
LiSLiTSunC, et al. WPConvNet: an interpretable wavelet packet kernel-constrained convolutional network for noise-robust fault diagnosis. IEEE Trans Neural Netw Learn Syst2023; 35(10): 14974–14988.
33.
FuGMaoKZhangG, et al. Attention mechanism feature fusion for bearing fault diagnosis in noise environment. In: Proceedings of the 2024 IEEE 3rd international conference on electrical engineering, big data and algorithms (EEBDA), Changchun, China, February 2024, pp. 420–425. IEEE.
34.
LeCunYBottouLBengioY, et al. Gradient-based learning applied to document recognition. Proc IEEE2002; 86(11): 2278–2324.
35.
SzegedyCLiuWJiaY, et al. Going deeper with convolutions. In: Proceedings of the 2015 IEEE/CVF conference on computer vision and pattern recognition (CVPR), IEEE, Boston, MA, 2015, pp. 1–9.
36.
PiccardiESGligaT.Understanding sensory regulation in typical and atypical development: the case of sensory seeking. Dev Rev2022; 65: 101037.
37.
WooSParkJLeeJY, et al. Cbam: convolutional block attention module. In: Proceedings of the 2018 European conference on computer vision (ECCV), Springer Nature, Munich, Germany, 2018, pp. 3–19.
38.
HuangHBaddourN.Bearing vibration data collected under time-varying rotational speed conditions. Data Brief2018; 21: 1745–1749.
39.
SmithWARandallRB.Rolling element bearing diagnostics using the Case Western Reserve University data: a benchmark study. Mech Syst Signal Process2015; 64: 100–131.
40.
YanSShaoHWangJ, et al. LiConvFormer: a lightweight fault diagnosis framework using separable multiscale convolution and broadcast self-attention. Expert Syst Appl2024; 237: 121338.
41.
SandlerMHowardAZhuM, et al. Mobilenetv2: inverted residuals and linear bottlenecks. In: Proceedings of the 2018 IEEE/CVF conference on computer vision and pattern recognition (CVPR), IEEE, Salt Lake City, UT,2018, pp. 4510–4520.
42.
YuCGaoCWangJ, et al. Bisenet v2: bilateral network with guided aggregation for real-time semantic segmentation. Int J Comput Vis2021; 129(11): 3051–3068.
43.
ZhaoMZhongSFuX, et al. Deep residual shrinkage networks for fault diagnosis. IEEE Trans Ind Inform2019; 16(7): 4681–4690.
44.
ChenBLiuTHeC, et al. Fault diagnosis for limited annotation signals and strong noise based on interpretable attention mechanism. IEEE Sens J2022; 22(12): 11865–11880.
45.
FanQHuangHChenM, et al. Rmt: retentive networks meet vision transformers. In: Proceedings of the 2024 IEEE/CVF conference on computer vision and pattern recognition (CVPR), IEEE, Seattle, WA, 2024, pp. 5641–5651.
46.
WangAChenHLinZ, et al. LSNet: see large, focus small. In: Proceedings of the 2025 IEEE/CVF conference on computer vision and pattern recognition conference (CVPR), IEEE, Nashville, TN, 2025, pp. 9718–9729.
47.
Van Der MaatenL. Accelerating t-SNE using tree-based algorithms. J Mach Learn Res2014; 15(1): 3221–3245.
48.
ZhangXHeCLuY, et al. Fault diagnosis for small samples based on attention mechanism. Measurement2022; 187: 110242.
49.
QinYLiuXLiX, et al. Simulation-data driven generalized zero-shot learning for multi-agent bearing compound fault diagnosis. Knowl Based Syst2025; 318: 113595.
50.
WangLChenDMaoY, et al. A zero-shot attribute-embedded model with a feature difference mapping sigmoid function for compound fault diagnosis of rotating machinery. ISA Trans2025; 157: 451–465.
