The diagnosis of bearing compound faults, particularly zero-shot compound fault diagnosis, has become a significant focus in recent research. While semantic learning methods have introduced novel approaches for addressing zero-shot diagnosis tasks, their diagnostic accuracy can be compromised by the mutual interference between single and compound faults in practical scenarios. To overcome these challenges, this article presents a generalized zero-shot compound fault diagnosis method that integrates a reserved gate classifier and physical information-constrained semantics. By effectively distinguishing single faults from compound faults and implementing tailored diagnostic strategies, this method achieves robust and accurate fault diagnosis under generalized zero-sample conditions. First, a reserved gate pre-classifier (RGPC) and a global-local feature extractor (GLFE) are trained using interpolation strategies to classify both single and compound faults. Physical information from single faults is leveraged to construct a physical information-constrained autoencoder (PICAE), which generates high-fidelity semantic representations. Finally, samples identified as compound faults by the reserved gate outputs are mapped into the semantic space using global-local fusion features. The compound fault types are then determined by comparing the mapped features with predefined compound fault semantics. The proposed method was validated on a bearing compound fault dataset, achieving diagnostic accuracies of 86.71% for three compound fault types and 78.63% for four compound fault types.
HuangWMaJQiuT, et al. Iteratively reweighted accurate sparse low-rank matrix estimation algorithm for bearing fault diagnosis. Measurement2024; 225: 113959.
2.
YeMYanXHuaX, et al. MRCFN: a multi-sensor residual convolutional fusion network for intelligent fault diagnosis of bearings in noisy and small sample scenarios. Expert Syst Appl2025; 259: 125214.
3.
ChenYTaoLLiuX, et al. Open-set fault recognition and inference for rolling bearing based on open fault semantic subspace. IEEE Trans Instrum Meas2024; 73: 1–11.
4.
DongWWuXZhangS, et al. An early fault warning and diagnosis model based on a two-step analysis strategy for wind turbine bearings. Struct Health Monit2024; 14759217241270978.
5.
WangXXuGZhouZ, et al. Sequential feature-augmented deep multilabel learning for compound fault diagnosis of rotating machinery with few labeled and imbalanced data. IEEE Trans Ind Inform2024; 20: 13947–13955.
6.
HeYZhaoCShenW.Cross-domain compound fault diagnosis of machine-level motors via time–frequency self-contrastive learning. IEEE Trans Ind Inform2024; 20: 9692–9701.
7.
HuangRXiaJZhangB, et al. Compound fault diagnosis for rotating machinery: State-of-the-art, challenges, and opportunities. J Dyn Monit Diagn2023; 2(1): 13–29.
8.
LiWLanHChenJ, et al. WavCapsNet: an interpretable intelligent compound fault diagnosis method by backward tracking. IEEE Trans Instrum Meas2023; 72: 1–11.
9.
HuangRLiJLiaoY, et al. Deep adversarial capsule network for compound fault diagnosis of machinery toward multidomain generalization task. IEEE Trans Instrum Meas2021; 70: 1–11.
10.
HuangRLiJLiW, et al. Deep ensemble capsule network for intelligent compound fault diagnosis using multisensory data. IEEE Trans Instrum Meas2020; 69: 2304–2314.
11.
MaPZhangHWangC.Adaptive dynamic mode decomposition and its application in rolling bearing compound fault diagnosis. Struct Health Monit2023; 22: 398–416.
12.
XuYFengGTangX, et al. A Modulation signal bispectrum enhanced squared envelope for the detection and diagnosis of compound epicyclic gear faults. Struct Health Monit2023; 22: 562–580.
13.
FanCZhangYMaH, et al. A novel metric-based model with the ability of zero-shot learning for intelligent fault diagnosis. Eng Appl Artif Intell2024; 129: 107605.
14.
ZengYDengCXiongF.Enhancing rubber rupture detection in rubber bearing through generative adversarial network and feature-bagging zero-shot methodology. Struct Health Monit2024; 24(3): 1824–1840.
15.
XingSLeiYWangS, et al. A label description space embedded model for zero-shot intelligent diagnosis of mechanical compound faults. Mech Syst Signal Process2022; 162: 108036.
16.
WangXZhangLWangY, et al. Visual-Language contrastive learning for zero-shot compound fault diagnosis in sucker rod wells. Measurement2025; 243: 116320.
17.
YangQLinYRuanJ.A zero-shot fault attribute transfer learning method for compound fault diagnosis of power circuit breakers. Meas Sci Technol2024; 35: 056111.
18.
QinNYinYHuangD, et al. Generalized zero-shot learning for fault diagnosis in high-speed train bogies based on enhanced diffusion generative models. IEEE Transactions on Reliability2024; 1–13.
19.
WangJXuZNiuF, et al. Adaptively weighted semantic autoencoder for zero-shot compound fault diagnosis. IEEE Sensors J2024; 24: 37472–37481.
20.
CaiLYinHLinJ, et al. Federated generalized zero-sample industrial fault diagnosis across multisource domains. IEEE Internet Things J2024; 11: 38895–38906.
21.
MaCWangXLiY, et al. Broad zero-shot diagnosis for rotating machinery with untrained compound faults. Reliability Eng Syst Saf2024; 242: 109704.
22.
CaiLDengXYinH, et al. Generalized zero-sample industrial fault diagnosis with domain bias. Reliability Eng Syst Saf2025; 253: 110571.
23.
KimHKShimJ.Generalized zero-shot learning for classifying unseen wafer map patterns. Eng Appl Artif Intell2024; 133: 108476.
XuJLiangSDingX, et al. A zero-shot fault semantics learning model for compound fault diagnosis. Expert Syst Appl2023; 221: 119642.
26.
XuJZhangHZhouL, et al. Zero-shot compound fault diagnosis method based on semantic learning and discriminative features. IEEE Trans Instrum Meas2023; 72: 1–13.
27.
ChenHHongJChengC, et al. Zero-shot compound fault diagnosis method using semantic construction and embedding. IEEE Trans Instrum Meas2024; 73: 1–13.
28.
ZhangXHuangWDingC, et al. Cross-supervised multisource prototypical network: a novel domain adaptation method for multi-source few-shot fault diagnosis. Adv Eng Inf2024; 61: 102538.
29.
WangJRenHShenC, et al. Multi-scale style generative and adversarial contrastive networks for single domain generalization fault diagnosis. Reliability Eng Syst Saf2024; 243: 109879.
30.
OhBKGlisicBKimY, et al. Convolutional neural network–based data recovery method for structural health monitoring. Struct Health Monit2020; 19: 1821–1838.
31.
YangRWangRDengY, et al. Rethinking the random cropping data augmentation method used in the training of CNN-based SAR image ship detector. Remote Sens2021; 13: 34.
32.
HuangWDengYHuiS, et al. Sparse self-attention transformer for image inpainting. Pattern Recognit2024; 145: 109897.
33.
ShiJWangYYuZ, et al. Exploiting multi-scale parallel self-attention and local variation via dual-branch transformer-CNN structure for face super-resolution. IEEE Trans Multimedia2024; 26: 2608–2620.
34.
ZhaoWWangLZhangZ, et al. Quadratic interpolation optimization (QIO): a new optimization algorithm based on generalized quadratic interpolation and its applications to real-world engineering problems. Comput Methods Appl Mech Eng2023; 417: 116446.
35.
TengSLiuZLuoW, et al. Bridge anomaly detection based on reconstruction error and structural similarity of unsupervised convolutional auto-encoder. Struct Health Monit2024; 23: 2221–2237.
36.
BenestyJChenJHuangY, et al. Pearson correlation coefficient. In: CohenIHuangYChenJ, et al. (eds.) Noise Reduction in Speech Processing. Berlin, Heidelberg: Springer, 2009, pp. 1–4.
37.
RahadianHBandongSWidyotriatmoA, et al. Image encoding selection based on Pearson correlation coefficient for time series anomaly detection. Alexandria Eng J2023; 82: 304–322.
38.
CenJZhaoBLiuX, et al. An intelligent compound fault diagnosis method using generalized zero-shot model of bearing. Meas Sci Technol2024; 35: 096134.
39.
ZhangYQinNHuangD, et al. Generalized zero-shot approach leveraging attribute space for high-speed train bogie. IEEE Trans Instrum Meas2024; 73: 1–12.
40.
XuJZhangHChenW, et al. CGASNet: a generalized zero-shot learning compound fault diagnosis approach for bearings. IEEE Trans Instrum Meas2024; 73: 1–11.
41.
ShinJ-SKimM-JKimB-S, et al. Enhanced detection of unknown defect patterns on wafer bin maps based on an open-set recognition approach. Comput Ind2025; 164: 104208.
42.
JiangMXiangYShengC.Based on spatial Mahalanobis distance: a novel zero-shot learning method for compound fault identification and decoupling. Expert Syst Appl2025; 272: 126680.
