Addressing the issues of limited fault samples for piston pumps, weak fault signals under noise interference, and traditional deep learning's heavy reliance on vast amounts of training data, we proposed a novel few-shot fault diagnosis approach for piston pumps based on model-agnostic meta-learning (MAML) with few samples. Firstly, the collected one-dimensional vibration signal was decomposed using improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), resulting in intrinsic mode function (IMF) components. Subsequently, sensitive components rich in fault information were selected to enhance the feature information within the vibration signal. Secondly, a multi-channel one-dimensional convolutional model was established, incorporating a channel interaction feature encoder equipped with an efficient channel attention mechanism. This design aimed to focus on the mutual fault information among different channels, thereby effectively extracting general diagnostic knowledge applicable to multiple diagnostic meta-tasks. Finally, the one-dimensional convolutional model served as the base model, which was trained through the MAML method to obtain optimal initial model parameters. Following this, the optimally initialized model could quickly adapt to new operating conditions with limited piston pump fault samples, thereby realizing few-shot fault diagnosis for piston pumps. The performance of the proposed model was validated using experimental data from piston pump tests. Experimental results demonstrate that the proposed method achieves an accuracy rate of over 90% across various diagnostic tasks under few-sample conditions.
LIChunlin, XIONGJianbin, SUNaiquan, et al. Application review of deep learning in fault diagnosis[J]. Machine Tools and Hydraulics, 2020, 48(13): 174-184. (in Chinese)
[5]
ZHANGTianci, CHENJinglong, LIFudong, et al. Intelligent fault diagnosis of machines with small & imbalanced data: A state-of-the-art review and possible extensions, ISA Transactions[J]. 2022, 119: 152-171.
[6]
FINNC, ABBEELP, LEVINES. Model-agnostic meta-learning for fast adaptation of deep networks[C]//International Conference on Machine Learning, PMLR, 2017: 1126-1135.
[7]
HOUX, XUJ, WUJ, et al. Cross domain adaptation of crowd counting with model-agnostic meta-learning[J]. Applied Sciences, 2021, 11(24): 12037.
[8]
KANGJ W, LIUR Q, LIL T, et al. Domain-invariant speaker vector projection by model-agnostic meta-learning[DB/OL]. (2020-05-25)[2024-01-01].
ZHAOZ B, LIT F, WUJ Y, et al. Deep learning algorithms for rotating machinery intelligent diagnosis: An open source benchmark study[J]. ISA Transactions, 2020, 107(10): 224-255.
[11]
LINJ, SHAOH, ZHOUX, et al. Generalized MAML for few-shot cross-domain fault diagnosis of bearing driven by heterogeneous signals[J]. Expert Systems with Applications, 2023, 230: 120696.
[12]
LUOJ, SHAOH, LINJ, et al. Meta-learning with elastic prototypical network for fault transfer diagnosis of bearings under unstable speeds[J]. Reliability Engineering & System Safety, 2024, 245: 110001.
YANBingsheng, LIUZhaoliang, LIUZirang. Research on fault diagnosis method of cross meachanical components based on meta learning under small sample[J]. Combined Machine Tools and Automated Processing Technology, 2022(10): 136-140. (in Chinese)
[15]
LIUS, HUANGJ, MAJ, et al. Class-incremental continual learning model for plunger pump faults based on weight space meta-representation[J]. Mechanical Systems and Signal Processing, 2023, 196: 110309.
LUXinxin, MAJun, ZHANGYingcong. Fault diagnosis of small sample automobile planetary gearboxes based on continuous wavelet transform and model agnostic meta learning[J]. Journal of Mechanical Transmission, 2022, 46(9): 159-164. (in Chinese)
[18]
LIC, LIS, WANGH, et al. Attention-based deep meta-transfer learning for few-shot fine-grained fault diagnosis[J]. Knowledge-Based Systems, 2023, 264: 110345.
YANGQing, DONGYansong, WUDongsheng, et al. Bearing fault diagnosis under variable working conditions based on MTLSAM model under small sample conditions[J]. Combined Machine Tools and Automated Processing Technology, 2023(2): 78-81. (in Chinese)
ZHUJuntao, YAOGuangle, ZHANGGexiang, et al. Survey of few shot learning of deep neural network[J]. Computer Engineering and Applications, 2021, 57(7): 22-33. (in Chinese)
[23]
HUJ, SHENL, SUNG. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[24]
ZHAOM, ZHONGS, FUX, et al. Deep residual shrinkage networks for fault diagnosis[J]. IEEE Transactions on Industrial Informatics, 2019, 16(7): 4681-4690.
LIUShengzheng, ZHANGLin, ZENGXianghui, et al. Loose sliper fault diagbosis of axial piston pump based on MED-EEMD and ELM.[J]. Journal of Mechanical & Electrical Engineering, 2020, 37(3): 241-246. (in Chinese)
MAHongwei, ZHANGDawei, CAOXiangang, et al. Research on vibration signal denoising method based on EMD[J]. Journal of Vibration and Shock, 2016, 35(22): 38-40. (in Chinese)