Mechanical fault diagnosis is crucial for maintaining the stability of industrial systems. However, the limited labeled fault data in real-world applications severely restricts diagnostic accuracy and generalization. To address this problem, this paper proposes a deep reinforcement learning–enhanced multi-view generative synergistic optimization agent, termed MGSOA, for machinery fault diagnosis under data-scarce conditions. First, a deep data sampling augmented generative adversarial network is designed to generate high-quality fault samples and mitigate data bias in few-shot scenarios. Furthermore, a multi-view feature extraction with domain adaptation architecture is constructed to adaptively extract multi-scale and multi-view features, enhancing feature representation under few-shot and mixed operating conditions. Finally, a novel priority-embedded reward-centering deep deterministic policy gradient algorithm is developed to improve the accuracy and robustness of the agent. Experimental results demonstrate that the proposed method outperforms other state-of-the-art methods in few-shot machinery fault diagnosis.
QinGZhangKLaiX, et al. An adaptive symmetric loss in dynamic wide-kernel resnet for rotating machinery fault diagnosis under noisy labels. IEEE Trans Instrum Meas2024; 73: 3517512.
2.
LuoSZhangDWuJ, et al. A limited annotated sample fault diagnosis algorithm based on nonlinear coupling self-attention mechanism. Eng Fail Anal2025; 174: 109474.
3.
LiuDCuiLWangG, et al. Interpretable domain adaptation transformer: a transfer learning method for fault diagnosis of rotating machinery. Struct Health Monit2025; 24(2): 1187–1200.
4.
YangXYeTYuanX, et al. A novel data augmentation method based on denoising diffusion probabilistic model for fault diagnosis under imbalanced data. IEEE Trans Ind Informat2024; 20(5): 7820–7831.
5.
TianJJiangYZhangJ, et al. A novel data augmentation approach to fault diagnosis with class-imbalance problem. Reliab Eng Syst Saf2024; 243: 109832.
6.
LiuCHeSLiS, et al. Time-series multi-instance learning for weakly supervised industrial fault detection. IEEE Trans Ind Informat2025; 21(4): 3326–3335.
7.
ChenMShaoHDouH, et al. Data augmentation and intelligent fault diagnosis of planetary gearbox using ilofgan under extremely limited samples. IEEE Trans Rel2023; 72(3): 1029–1037.
8.
FanJYuanXMiaoZ, et al. Full attention wasserstein gan with gradient normalization for fault diagnosis under imbalanced data. IEEE Trans Instrum Meas2022; 71: 3517516.
9.
LyuPChengYZhangM, et al. GPSC-GAN: a data enhanced model for intelligent fault diagnosis. IEEE Trans Instrum Meas2024; 73: 3532116.
10.
GuoQLiYSongY, et al. Intelligent fault diagnosis method based on full 1-D convolutional generative adversarial network. IEEE Trans Ind Informat2019: 16(3): 2044–2053.
11.
PuZYanLBaiY, et al. Incrementally generative adversarial diagnostics using few-shot enabled one-class learning. IEEE Trans Ind Informat2024; 20(10): 12189–12199.
12.
Maldonado-CorreaJTorres-CabreraJMartín-MartínezS, et al. Wind turbine fault detection based on the transformer model using scada data. Eng Fail Anal2024; 162: 108354.
13.
SuYMengLKongX, et al. Small sample fault diagnosis method for wind turbine gearbox based on optimized generative adversarial networks. Eng Fail Anal2022; 140: 106573.
14.
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.
15.
ZhangLQiuJLiuJ, et al. CTNet: an interpretable convolutional neural network for bearing fault diagnosis. Struct Health Monit. Epub ahead of print 7October2025. DOI: 10.1177/14759217251382151.
16.
WangYPanZYuanX, et al. A novel deep learning based fault diagnosis approach for chemical process with extended deep belief network. ISA Trans2020; 96: 457–467.
17.
ChenZJiJCNiQ, et al. Bi-structural spatial–temporal network for few-shot fault diagnosis of rotating machinery. Mech Syst Signal Process2025; 227: 112378.
18.
YuYJiJCChenZ, et al. Large language model to assist data augmentation in soft contrastive learning for few-shot machinery fault diagnosis. Adv Eng Inform2026; 69: 104078.
19.
LiGWuJDengC, et al. Deep reinforcement learning-based online domain adaptation method for fault diagnosis of rotating machinery. IEEE/ASME Trans Mech2021; 27(5): 2796–2805.
20.
SunJFeiXXuZ, et al. Dynamic wavelet optimization and hierarchical reinforcement learning for fault diagnosis in wind turbines under variable operating conditions. Nondestr Test Eval. Epub ahead of print 28July2025. DOI: 10.1080/10589759.2025.2539482.
21.
LiuXZhaoZYangF, et al. Environment adaptive deep reinforcement learning for intelligent fault diagnosis. Eng Appl Artif Intell2025; 151: 110783.
22.
LuRJiangZYangT, et al. A novel hybrid-action-based deep reinforcement learning for industrial energy management. IEEE Trans Ind Informat2024; 20(10): 12461–12475.
23.
QinPZhaoT. Knowledge guided deep deterministic policy gradient. Knowl Based Syst2025; 311: 113087.
24.
HanSZhouWLüS, et al. Regularly updated deterministic policy gradient algorithm. Knowl Based Syst2021; 214: 106736.
25.
ZhouDHuangDHaoJ, et al. Vibration-based fault diagnosis of the natural gas compressor using adaptive stochastic resonance realized by generative adversarial networks. Eng Fail Anal2020; 116: 104759.
26.
ZhouTLiQLuH, et al. GAN review: models and medical image fusion applications. Inf Fusion2023; 91: 134–148.
27.
WangZXuanJ. Intelligent fault recognition framework by using deep reinforcement learning with one dimension convolution and improved actor-critic algorithm. Adv Eng Inform2021; 49: 101315.
28.
LuoMXuJFanY, et al. TRNet: a cross-component few-shot mechanical fault diagnosis. IEEE Trans Ind Informat2022; 19(5): 6883–6894.
29.
ChenZPengLFanJ, et al. Epbs_fidmv: a fault injection and diagnosis methods validation benchmark for epbs of emu. Control Eng Pract2024; 145: 105873.
30.
WangCTianBYangJ, et al. Neural-transformer: a brain-inspired lightweight mechanical fault diagnosis method under noise. Reliab Eng Syst Saf2024; 251: 110409.
31.
SongXCongYSongY, et al. A bearing fault diagnosis model based on cnn with wide convolution kernels. J Ambient Intell Hum Comput2021; 13(8): 4041–4056.
32.
WangCSunHCaoX. Construction of the efficient attention prototypical net based on the time–frequency characterization of vibration signals under noisy small sample. Measurement2021; 179: 109412.
33.
SunHWangCCaoX. An adaptive anti-noise gear fault diagnosis method based on attention residual prototypical network under limited samples. Appl Soft Comput2022; 125: 109120.
34.
ZhangJYiSLiangG, et al. A new bearing fault diagnosis method based on modified convolutional neural networks. Chinese J Aeronaut2020; 33(2): 439–447.
35.
ZhangXHeCLuY, et al. Fault diagnosis for small samples based on attention mechanism. Measurement2022; 187: 110242.
36.
FangQWuD. ANS-net: anti-noise siamese network for bearing fault diagnosis with a few data. Nonlinear Dyn2021; 104: 2497–2514.
37.
HouYWangJChenZ, et al. Diagnosisformer: an efficient rolling bearing fault diagnosis method based on improved transformer. Eng Appl Artif Intell2023; 124: 106507.
38.
WangCXuMZhangQ, et al. A novel lightweight rotating mechanical fault diagnosis framework with adaptive residual enhancement and multi-group coordinate attention. IEEE Trans Instrum Meas2025; 74: 3514517.
39.
WuWZhouNLiangX, et al. TCAC-transformer: a fast convolutional transformer with temporal-channel attention for efficient industrial fault diagnosis. Expert Syst Appl2025; 297: 129473.
40.
HeDZhangZJinZ, et al. RTSMFFDE-HKRR: a fault diagnosis method for train bearing in noise environment. Measurement2025; 239: 115417.
