Existing fault diagnosis methods primarily focus on directly extracting fault-sensitive (highly discriminative) features from rotating machinery signals, while nonfault-sensitive (low-discriminative) features—typically dominated by background information—often prevail in the spectral background and exert a significant influence on overall signal representations. This dominance makes fault-related features difficult to be robustly extracted and modeled under multinoise and complex operating conditions. In addition, single-modality features are insufficient to comprehensively characterize the overall operating states of mechanical equipment, which further limits the robustness and cross-scenario generalization capability of fault diagnosis models. To address these challenges, this article proposes an LCR-Net that integrates regional bidirectional cross-attention (RBCA) and causal inference (CI). Specifically, the RBCA mechanism enables bidirectional cross-modal information interaction between acoustic and vibration signals at the level of low-discriminative features, facilitating their collaborative enhancement to serve as a stable background reference. Meanwhile, residual learning is employed to model residual features associated with fault evolution. Subsequently, a CI module is introduced to estimate causal weights across multimodal signals, thereby alleviating the instability of early fusion strategies under complex noise conditions and jointly achieving causality-driven residual feature weighted fusion and fault identification. To validate the effectiveness of the proposed method, an experimental dataset containing multiple noise types with different noise intensities is constructed. Experimental results demonstrate that LCR-Net achieves an average diagnostic accuracy of 99.68% under different train–test split ratios within only 30 training epochs, significantly outperforming diagnostic methods based on single-modality acoustic or vibration signals. Furthermore, in cross-device generalization experiments conducted on bearing and motor datasets, the proposed method attains average diagnostic accuracies of 99.3% and 98.2%, respectively. These results indicate that the proposed approach exhibits strong robustness and generalization capability for multimodal fault diagnosis under multinoise and cross-device scenarios.
GawdeSPatilSKumarS, et al. Multi-fault diagnosis of industrial rotating machines using data-driven approach: a review of two decades of research. Eng Appl Artif Intell2023; 123: 106139.
2.
ZastepaM. Commutator motor fault diagnosis using acoustic data with a transfer learning approach. Prz Elektrotech2024; 12: 173–180.
3.
ZhouPChenSHeQ, et al. Rotating machinery fault-induced vibration signal modulation effects: a review with mechanisms, extraction methods and applications for diagnosis. Mech Syst Signal Process2023; 200: 110489.
4.
ZhouQMaWZhangY, et al. Bearing fault diagnosis for variable working conditions via lightweight transformer and homogeneous generalized contrastive learning with inter-class repulsive discriminant. Eng Appl Artif Intell2025; 139: 109548.
5.
BadihiHZhangYJiangB, et al. A comprehensive review on signal-based and model-based condition monitoring of wind turbines: fault diagnosis and lifetime prognosis. Proceedings of the IEEE2022; 110(6): 754–806.
6.
YanRShangZXuH, et al. Wavelet transform for rotary machine fault diagnosis: 10 years revisited. Mech Syst Signal Process2023; 200: 110545.
7.
Mohammad-AlikhaniAJamshidpourEDhaleS, et al. Fault diagnosis of electric motors by a channel-wise regulated CNN and differential of STFT. IEEE Trans Ind Appl2025; 61: 3066–3077.
8.
ShanSLiuJWuS, et al. A motor bearing fault voiceprint recognition method based on Mel-CNN model. Measurement2023; 207: 112408.
9.
WangYLiDLiL, et al. A novel deep learning framework for rolling bearing fault diagnosis enhancement using VAE-augmented CNN model. Heliyon2024; 10(15): e35407.
10.
LiuCTongJZhengJ, et al. Rolling bearing fault diagnosis method based on multi-sensor two-stage fusion. Meas Sci Technol2022; 33(12): 125105.
11.
SunDLiYLiuZ, et al. Physics-inspired multimodal machine learning for adaptive correlation fusion based rotating machinery fault diagnosis. Inform Fusion2024; 108: 102394.
12.
PengCShengYGuiW, et al. A rolling bearing fault diagnosis method based on multimodal knowledge graph. IEEE Trans Ind Inform2024; 20: 13047–13057.
13.
DingSZhouGWangX, et al. Fault diagnosis of wind turbine rotating bearing based on multi-mode signal enhancement and fusion. Entropy2025; 27(9): 951.
14.
YouKLiuCLinY, et al. DTMPI-DIVR: a digital twins for multi-margin physical information via dynamic interaction of virtual and real sound-vibration signals for bearing fault diagnosis without real fault samples. Expert Syst Appl2025; 292: 128592.
15.
ShiHLiYBaiX, et al. A two-stage sound-vibration signal fusion method for weak fault detection in rolling bearing systems. Mech Syst Signal Process2022; 172: 109012.
16.
YouKWangPHuangP, et al. A sound-vibration physical-information fusion constraint-guided deep learning method for rolling bearing fault diagnosis. Reliab Eng Syst Saf2025; 253: 110556.
17.
WangQSunZZhuY, et al. Intelligent fault diagnosis algorithm of rolling bearing based on optimization algorithm fusion convolutional neural network. Electron Res Arch2023; 20(11): 19963–19982.
18.
HouWZhangCJiangY, et al. A new bearing fault diagnosis method via simulation data driving transfer learning without target fault data. Measurement2023; 215: 112879.
19.
FuZLiuZPingS, et al. TRA-ACGAN: a motor bearing fault diagnosis model based on an auxiliary classifier generative adversarial network and transformer network. ISA Trans2024; 149: 381–393.
20.
MengLXieJChengL, et al. Multi-condition bearing fault diagnosis method based on RP–MBRN. Struct Health Monit2024.
21.
KimHParkCHSuhC, et al. MPARN: multi-scale path attention residual network for fault diagnosis of rotating machines. J Comput Des Eng2023; 10(2): 860–872.
22.
DengLZhangYZhaoC, et al. Dual-path multi-scale attention residual network for fault diagnosis of rolling bearings under complex operating conditions. Meas Sci Technol2024; 35(8): 086106.
23.
MengLSuYKongX, et al. Intelligent fault diagnosis of gearbox based on differential continuous wavelet transform-parallel multi-block fusion residual network. Measurement2023; 206: 112318.
24.
ShiJRenYTangH, et al. Hydraulic directional valve fault diagnosis using a weighted adaptive fusion of multi-dimensional features of a multi-sensor. J Zhejiang Univ Sci A2022; 23(4): 257–271.
25.
ZhangTHeSChenJ, et al. Toward small sample challenge in intelligent fault diagnosis: attention-weighted multidepth feature fusion net with signals augmentation. IEEE Trans Instrum Meas2021; 71: 1–13.
26.
JiaSLiYWangX, et al. Deep causal factorization network: a novel domain generalization method for cross-machine bearing fault diagnosis. Mech Syst Signal Process2023; 192: 110228.
27.
WeiHGuoPLiQ. Application of the causal graph-based fault inference method in reliability analysis. In: YanLDuanHDengY (eds). Advances in guidance, navigation and control. Singapore: Springer Nature Singapore, 2023, pp. 3715–3727.
28.
AhsanMSalahMM. Similarity index of the STFT-based health diagnosis of variable speed rotating machines. Intell Syst Appl2023; 20: 200270.
29.
SunBShengZSongP, et al. State-of-the-art detection and diagnosis methods for rolling bearing defects: a comprehensive review. Appl Sci2025; 15(2): 1001.
30.
AkbalıkFYıldızAErtuğrulÖF, et al. Engine fault detection by sound analysis and machine learning. Appl Sci2024; 14(15): 6532.
31.
ZhaoYZhangNZhangZ, et al. Bearing fault diagnosis based on mel frequency cepstrum coefficient and deformable space-frequency attention network. IEEE Access2023; 11: 34407–34420.
32.
JungWKimSYunS, et al. Vibration, acoustic, temperature, and motor current dataset of rotating machine under varying operating conditions for fault diagnosis. Data Brief2023; 48: 109049.
33.
SehriMDumondPBouchardM. University of Ottawa constant load and speed rolling-element bearing vibration and acoustic fault signature datasets. Data Brief2023; 49: 109327.
34.
SehriMDumondP. University of Ottawa constant and variable speed electric motor vibration and acoustic fault signature dataset. Data Brief2024; 53: 110144.
35.
MaZChenYZhangT, et al. An enhanced spectral amplitude modulation method for fault diagnosis of rolling bearings. Machines2024; 12(11): 779.
36.
BingHChenBXiangL, et al. Dynamic simulation and experimental study of dual-rotor system induced by rub-impact fault with variable state and parameter. Eng Res Express2025; 7(1): 015353.
37.
AhmedSKopsaftopoulosF. A functional series time-dependent framework for non-stationary modeling and statistical damage diagnosis via ultrasonic guided waves. Struct Health Monit2025.
38.
AhmedSKopsaftopoulosF. Active sensing ultrasonic guided wave-based damage diagnosis via stochastic stationary time-series models. Struct Health Monit2024; 23(4): 2559–2595.
39.
RandallRBAntoniJ. Rolling element bearing diagnostics—a tutorial. Mech Syst Signal Process2011; 25(2): 485–520.
40.
DingXWuHWangJ, et al. Causal discovery for rolling bearing fault under missing data: from the perspective of causal effect and information flow. IEEE Trans Instrum Meas2024; 74: 1–17.
41.
JiaFWangYShenJ, et al. Stepwise feature norm network with adaptive weighting for open set cross-domain intelligent fault diagnosis of bearings. Meas Sci Technol2024; 35(5): 056126.
42.
LiYMengXXuF, et al. Intelligent fault diagnosis of hoisting systems under complex working conditions using deep graph convolutional generative adversarial networks with limited data. Struct Health Monit2025; 24(6): 3770–3791.
43.
RanGLiXLiM, et al. FD-Zero framework: LLM-based diagnosis framework for zero-shot mechanical time-domain signals. Struct Health Monit2025.
44.
ZhangSSongBZhaoF, et al. A decoupling method for compound faults of rolling bearings based on multi-modal bidirectional cross-attention fusion and contrastive learning. Meas Sci Technol2025; 36(10): 106131.
45.
AmerAKopsaftopoulosF. Gaussian process regression for active sensing probabilistic structural health monitoring: experimental assessment across multiple damage and loading scenarios. Struct Health Monit2023; 22(2): 1105–1139.
46.
SongEZhuRYaoC, et al. Generalized zero-shot bearing compound fault diagnosis based on reserved gate pre-classifier and physical information semantics. Struct Health Monit2025.
47.
KoTParkJJangJ, et al. Adaptive bearing fault diagnosis using reference frequency and modified scalogram. Struct Health Monit2024; 23(5): 2781–2794.
48.
ZhangXDelphaCDialloD. Performance evaluation of Jensen–Shannon divergence-based incipient fault diagnosis: theoretical proofs and validations. Struct Health Monit2023; 22(3): 1628–1646.
49.
ChenAKopsaftopoulosFMishraS. Online fault detection for metal additive manufacturing with data-driven time series models. In: BlaschEDaremaFMetaxasD (eds.) Dynamic data driven applications systems. Cham: Springer Nature Switzerland, 2026, pp. 23–30.
50.
SahaDHoqueMEChowdhuryMEH. Enhancing bearing fault diagnosis using transfer learning and random forest classification: a comparative study on variable working conditions. IEEE Access2023; 12: 5986–6000.
ZhaoLLiuYZhangZ, et al. Nonlinear convolutional sparse filtering with Gaussian fitting for rolling bearings compound fault diagnosis. Struct Health Monit2024.
53.
LiaoYLiWLianG, et al. Heterogeneous multi-sensor fusion for ac motor fault diagnosis via graph neural networks. Electronics2025; 14(10): 2005.
54.
MejbelBGSarowSAAl-SharifyMT, et al. A data fusion analysis and random forest learning for enhanced control and failure diagnosis in rotating machinery. J Fail Anal Prev2024; 24(6): 2979–2989.
55.
SuiTFengYSuiS, et al. A bearing fault diagnosis method combining multi-source information and multi-domain information fusion. Machines2025; 13(4): 289.
56.
SunSXiaXZhouH. Bearing fault diagnosis under time-varying speeds with limited samples using frequency temporal series graph and graph generative classified adversarial networks. Neurocomputing2025; 647: 130613.
57.
WangXLiuZDaiM, et al. Time-shift denoising combined with DWT-enhanced condition domain adaptation for motor bearing fault diagnosis via current signals. IEEE Sens J2024; 24: 35019–35035.
58.
LiYLiuXHuJ, et al. Graph optimization algorithm enhanced by dual-scale spectral features with contrastive learning for robust bearing fault diagnosis. Knowl Based Syst2025; 315: 113275.
59.
FangXZhengJJiangB. A rolling bearing fault diagnosis method based on vibro-acoustic data fusion and fast Fourier transform (FFT). Int J Data Sci Analytics2025; 20(3): 2377–2386.
