针对风电机组内部滚动轴承的故障样本不足而易导致故障诊断精度下降的问题,提出一种基于深度卷积条件生成对抗网络（DC-CWGAN）的诊断方法。首先对振动信号进行连续小波变换（CWT）,构建对应的时频特征图集,以增强模型的故障特征捕捉能力;其次使用卷积结构替代条件生成对抗网络（CGAN）中的全连接层,并引入Wasserstein距离重构CGAN的损失函数,以提升DC-CWGAN中生成样本的质量,并提高网络训练过程中的稳定性;然后通过模型迁移策略的应用提高目标分类网络的泛化能力和计算效率。实验证明,所提方法能有效提高小样本问题下的轴承诊断准确率。
Abstract
For the accuracy decreasing in fault diagnosis derived from the samples insufficient of the rolling bearings in wind turbines, a fault diagnosis method based on deep convolutional-conditional Wasserstein generative adversarial network （DC-CWGAN） is proposed. Firstly, the continuous wavelet transform （CWT） is applied to the vibration signals to construct a dataset of time-frequency feature images. As a result, the feature capture ability for the fault model is increased. Secondly, the fully connected layers in the conditional generative adversarial networks （CGAN） are replaced by the convolution al structures. Followingly, the Wasserstein distance is introduced to reconstruct the CGAN loss function. Thus the quality of the generated samples and the stability of the DC-CWGAN training are both strengthened. Thirdly, with the application of the model transfer strategy, the generalization and the computational efficiency of the objective classification network are enhanced. Finally, it is demonstrated that the diagnosis accuracy of the rolling bearings under the small-sample condition is improved effectively by the proposed method.
关键词
风电机组 /
滚动轴承 /
故障诊断 /
小样本 /
模型迁移 /
条件生成对抗网络
Key words
wind turbines /
rolling bearings /
fault diagnosis /
small sample /
model transfer /
conditional generative adversarial network
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] WANG Z W, ZHANG Q H, XIONG J B, et al.Fault diagnosis of a rolling bearing using wavelet packet denoising and random forests[J]. IEEE sensors journal, 2017, 17(17): 5581-5588.
[2] 安文杰, 陈长征, 田淼, 等. 基于迁移学习的风电机组轴承故障诊断研究[J]. 太阳能学报, 2023, 44(6): 367-373.
AN W J, CHEN C Z, TIAN M, et al.Research on bearing fault diagnosis of wind turbines based on transfer learning[J]. Acta energiae solaris sinica, 2023, 44(6): 367-373.
[3] ZHANG X L, HAN P, XU L, et al.Research on bearing fault diagnosis of wind turbine gearbox based on 1DCNN-PSO-SVM[J]. IEEE access, 2020, 8: 192248-192258.
[4] MENG L, SU Y H, KONG X J, et al.A probabilistic Bayesian parallel deep learning framework for wind turbine bearing fault diagnosis[J]. Sensors, 2022, 22(19): 7644.
[5] LI X, ZHANG W, DING Q, et al.Intelligent rotating machinery fault diagnosis based on deep learning using data augmentation[J]. Journal of intelligent manufacturing, 2020, 31(2): 433-452.
[6] KHAN A, HWANG H, KIM H S.Synthetic data augmentation and deep learning for the fault diagnosis of rotating machines[J]. Mathematics, 2021, 9(18): 2336.
[7] 卢锦玲, 张祥国, 张伟, 等. 基于改进辅助分类生成对抗网络的风机主轴承故障诊断[J]. 电力系统自动化, 2021, 45(7): 148-154.
LU J L, ZHANG X G, ZHANG W, et al.Fault diagnosis of main bearing of wind turbine based on improved auxiliary classifier generative adversarial network[J]. Automation of electric power systems, 2021, 45(7): 148-154.
[8] SHAO S Y, WANG P, YAN R Q.Generative adversarial networks for data augmentation in machine fault diagnosis[J]. Computers in industry, 2019, 106: 85-93.
[9] 李俊卿, 胡晓东, 王罗, 等. 考虑数据不足和基于合作博弈模型融合的风电机组轴承故障诊断方法[J]. 太阳能学报, 2024, 45(1): 234-241.
LI J Q, HU X D, WANG L, et al.Bearing fault diagnosis method for wind turbine considering insufficient DATAAND based on cooperative game model fusion[J]. Acta energiae solaris sinica, 2024, 45(1): 234-241.
[10] GUO Q W, LI Y B, SONG Y, et al.Intelligent fault diagnosis method based on full 1-D convolutional generative adversarial network[J]. IEEE transactions on industrial informatics, 2020, 16(3): 2044-2053.
[11] ZHAO K, JIA F, SHAO H D.A novel conditional weighting transfer Wasserstein auto-encoder for rolling bearing fault diagnosis with multi-source domains[J]. Knowledge-based systems, 2023, 262: 110203.
[12] 王玉静, 夏林, 康守强, 等. 基于多源域异构模型迁移的滚动轴承故障诊断方法[J]. 振动与冲击, 2023, 42(24): 257-266.
WANG Y J, XIA L, KANG S Q, et al.A fault diagnosis method of rolling bearings based on multi-source domain heterogeneous model transfer[J]. Journal of vibration and shock, 2023, 42(24): 257-266.
[13] HENDRIKS J, DUMOND P, KNOX D A.Towards better benchmarking using the CWRU bearing fault dataset[J]. Mechanical systems and signal processing, 2022, 169: 108732.
[14] SHAO H D, XIA M, HAN G J, et al.Intelligent fault diagnosis of rotor-bearing system under varying working conditions with modified transfer convolutional neural network and thermal images[J]. IEEE transactions on industrial informatics, 2021, 17(5): 3488-3496.
[15] RUAN D W, SONG X Z, GÜHMANN C, et al. Collaborative optimization of CNN and GAN for bearing fault diagnosis under unbalanced datasets[J]. Lubricants, 2021, 9(10): 105.
[16] LI Y, YANG R, WANG H S.Unsupervised method based on adversarial domain adaptation for bearing fault diagnosis[J]. Applied sciences, 2023, 13(12): 7157.
[17] YANG B, LEI Y G, LI X, et al.Targeted transfer learning through distribution barycenter medium for intelligent fault diagnosis of machines with data decentralization[J]. Expert systems with applications, 2024, 244: 122997.
[18] SUN R X, LIU X Y, LIU S L, et al.A game theory enhanced domain adaptation network for mechanical fault diagnosis[J]. Measurement science and technology, 2022, 33(11): 115501.
[19] LUO J J, SHAO H D, CAO H R, et al.Modified DSAN for unsupervised cross-domain fault diagnosis of bearing under speed fluctuation[J]. Journal of manufacturing systems, 2022, 65(1): 180-191.
基金
天津市自然科学基金重点项目（23JCZDJC01140）
PDF(4470 KB)
