基于CWT和优化Swin Transformer的风电齿轮箱故障诊断方法

PDF(2891 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (15) : 200-208.

论文

周舟1，陈捷1,2，吴明明3

作者信息 +

Fault diagnosis method for wind power gearbox based on wavelet transform and optimized Swin Transformer

ZHOU Zhou1, CHEN Jie1,2, WU Mingming3

Author information +

文章历史 +

摘要

针对传统故障诊断方法在风电齿轮箱运行故障诊断应用上的不足，提出一种基于小波变换(Continuous Wavelet Transform,CWT)和优化 Swin Transformer 的风电齿轮箱故障诊断方法。该方法利用小波变换将风电齿轮箱振动信号转换为时频图；使用SuperMix数据增强算法对样本进行扩充；利用迁移学习技术将模型预训练参数用于训练和优化Swin Transformer模型；将训练完成的优化 Swin Transformer模型应用于风场实际运维数据进行对比验证，分类准确率达到99.67%。验证结果表明该方法能够有效的实现风电齿轮箱故障诊断，并提高模型的识别准确率。

Abstract

To address the shortcomings of traditional fault diagnosis method in the application of wind turbine gearbox operation status identification, a fault diagnosis method for wind turbine gearbox based on wavelet transform and optimization Swin Transformer is proposed. This method uses wavelet transform to convert the vibration signal of wind turbine gearbox into a time-frequency diagram. Use the SuperMix data augmentation algorithm to augment the sample; The Swin Transformer model is trained and optimized by pre-trained model parameters using transfer learning technology. The trained and optimized Swin Transformer model is applied to the actual operation and maintenance data of the wind farm for comparison and verification, and the classification accuracy reaches 99.67%. The verification results show that the proposed method can effectively identify the operating status of wind turbine gearbox and improve the recognition accuracy of the model.

导出引用

周舟1，陈捷1,2，吴明明3. 基于CWT和优化Swin Transformer的风电齿轮箱故障诊断方法[J]. 振动与冲击, 2024, 43(15): 200-208

ZHOU Zhou1, CHEN Jie1,2, WU Mingming3. Fault diagnosis method for wind power gearbox based on wavelet transform and optimized Swin Transformer[J]. Journal of Vibration and Shock, 2024, 43(15): 200-208

参考文献

[1] 刘长良,王梓齐.基于MSET和集成学习的风电机组齿轮箱故障预警[J].太阳能学报,2020,41(11):228-233. LIU Chang-liang,WANG Zi-qi. Wind turbine gearbox fault warning based on MSET and integrated learning[J]. Acta Energiae Solaris Sinica, 2020, 41(11): 228-233. [2] 刘潇波.基于深度学习的风电机组传动链故障智能诊断[D]. 北京：华北电力大学,2022. LIU Xiao-bo. Intelligent Fault Diagnosis of Wind Turbine Drivetrain based on Deep Learning[D]. Bei Jing：School of North China Electric Power University，2022. [3] 郑近德,潘海洋,戚晓利,等.基于改进经验小波变换的时频分析方法及其在滚动轴承故障诊断中的应用[J].电子学报,2018,(02):358-364. ZHEN Jin-de,PAN Hai-yang,QI Xiao-li, et al. Enhanced Empirical Wavelet Transform Based Time- Frequency Analysis and Its Application to Rolling Bearing Fault Diagnosis[J]. Chinese Journal of Electronics,2018,(02):358-364. [4] WANG Y T, XIE Y Z, FAN L S, et al. STMG: Swin transformer for multi-label image recognition with graph convolution network [J]. Neural Computing & Applications, 2022, 34(12): 10051-10063. [5] Zhang A H, Yu D L, Zhang Z Q. TLSCA-SVM Fault Diagnosis Optimization Method Based on Transfer Learning[J]. Processes, 2022,10(2). [6] Mounir N, Ouadi H, Jrhilifa I. Short-term electric load forecasting using an EMD-BI-LSTM approach for smart grid energy management system[J]. Energy and Buildings, 2023, 288: 113022. [7] Liu Z, Lin Y, Cao Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2021: 10012-10022. [8] 吕鑫栋, 李娇, 邓真楠,等. 基于改进Transformer的结构化图像超分辨网络[J]. 浙江大学学报(工学版), 2023, 57 (05): 865-874+910. LV Xin-dong, LI Jiao, DENG Zhen-Nan, et al. Structured image super-resolution network based on improved Transformer[J]. Journal of Zhejiang University: Engineering Science, 2023, 57 (05): 865-874+910. [9] 苗壮,赵昕昕,李阳,等. 基于Swin Transformer的深度有监督哈希图像检索方法[J]. 湖南大学学报(自然科学版), 2023, 50 (08): 62-71. MIAO Zhuang, ZHAO Xin-xin, LI Yang, et al. Deep Supervised Hashing Image Retrieval Method Based on Swin Transformer[J]. Journal of Hunan University: Natural Sciences, 2023, 50 (08): 62-71. [10] 钱坤, 李晨瑄, 陈美杉, 等. 基于改进Swin Transformer的舰船目标实例分割算法[J]. 系统工程与电子技术, 2023, 45 (10): 3049-3057. QIAN Kun, LI Chen-xuan, CHEN Mei-shan, et al. Ship target instance segmentation algorithm based on improved Swin Transformer[J]. Systems Engineering and Electronics, 2023, 45 (10): 3049-3057. [11] 邓学欣.开放式故障诊断构架及动态测试分析方法研究[D].天津.天津大学,2004. DENG Xue-xin. Research on Open Fault Diagnosis Framework and Dynamic Meseaurement Anlysis Methods[D].Tian Jin: Tianjin University,2004. [12] 黄盟,毕晓阳,杨晓,等.基于人工数据融合的柴油机故障数据增强方法[J].振动与冲击,2023,(13):278-286. HUANG Meng, BI Xiao-yang, YANG Xiao, et al. Diesel engine fault data augmentation method based on artificial data fusion[J]. Journal of vibration and shock,2023,(13):278-286. [13] Terrance DeVries and Graham W Taylor. Improved regularization of convolutional neural networks with cutout. arXiv preprint arXiv:1708.04552, 2017. [14] Dongyoon Han, Jiwhan Kim, and Junmo Kim. Deep pyramidal residual networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5927–5935, 2017. [15] Alex Krizhevsky, Geoffrey Hinton, et al. Learning multiple layers of features from tiny images. Technical report, Citeseer, 2009. [16] Yuji Tokozume, Yoshitaka Ushiku, and Tatsuya Harada. Between-class learning for image classification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 5486–5494, 2018. [17] Sangdoo Yun, Dongyoon Han, Seong Joon, et al. Cutmix: Regularization strategy to train strong classifiers with localizable features. arXiv preprintarXiv:1905.04899, 2019. [18] Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412,2017. [19] Dabouei A, Soleymani S, Taherkhani F, et al. Supermix: Supervising the mixing data augmentation[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2021: 13794-13803. [20] 苏浩,杨鑫,向玲,等.基于深度对比迁移学习的变工况下机械故障诊断[J].振动工程学报,2023,(03):845-853. SU Hao, YANG Xin, XIANG Lin, et al. Mechanical fault diagnosis under variable working conditions based on in-depth comparative transfer learning[J]. Journal of vibration engineering,2023,(03):845-853. [21] 温竹鹏,陈捷,刘连华,等.基于小波变换和优化CNN的风电齿轮箱故障诊断[J].浙江大学学报(工学版),2022,(06):1212-1219. WEN Zhu-peng, CHEN Jie, LIU Lian-hua, et al. Fault diagnosis of wind power gearbox based on wavelet transform and improved CNN[J]. Journal of Zhejiang University: Engineering Science,2022,(06):1212-1219.