针对在滚动轴承故障诊断领域中存在的故障样本较少,健康样本丰富所导致的故障类别失衡问题以及环境中存在噪声与人为噪声标签干扰等问题,提出了一种基于混合裁剪失衡数据增强与SwinNet网络相结合的故障诊断模型(SwinNet-MCIDA)。首先,借鉴图像分类数据增强方法,利用混合裁剪失衡数据增强算法对失衡类别的数据进行裁剪、混合处理生成新的故障样本来增加样本量,构造出增强数据集,然后对增强数据集进行小波变换转换成时频图像,将所得图像输入到卷积神经网络与Swin Transformer编码器相结合的SwinNet网络模型中,进行特征提取和故障分类,从而实现滚动轴承故障的高效诊断。实验结果表明,本文所提出的SwinNet-MCIDA故障诊断方法不仅可以很好地解决滚动轴承故障诊断领域故障类别失衡问题,而且也可以很好地应对故障数据中存在环境噪声问题与人为噪声标签干扰问题。
Abstract
In the field of fault diagnosis for rolling bearings, this study addresses the challenges of limited fault samples and imbalanced fault categories caused by an abundance of healthy samples. Furthermore, it tackles the issues of noise and artificial label interference present in the environment. To overcome these challenges, a fault diagnosis model named SwinNet-MCIDA, combining hybrid imbalance data augmentation and the SwinNet network, is proposed.Firstly, drawing inspiration from image classification data augmentation techniques, the MCIDA algorithm is employed to crop and mix imbalanced data, generating new fault samples to augment the dataset. The enhanced dataset is then transformed into time-frequency images using wavelet transform. These images are fed into the SwinNet network model, which combines a convolutional neural network and a Swin Transformer encoder, for feature extraction and fault classification, enabling efficient diagnosis of rolling bearing faults. The experimental results demonstrate that the proposed SwinNet-MCIDA fault diagnosis method not only effectively addresses the issue of imbalanced fault categories in the field of rolling bearing fault diagnosis, but also effectively handles the presence of environmental noise and artificial label interference in fault data.
关键词
滚动轴承 /
故障诊断 /
数据增强 /
卷积神经网络 /
Swin Transformer
{{custom_keyword}} /
Key words
rolling bearing /
fault diagnosis /
data augmentation /
convolutional neural network(CNN) /
Swin Transformer
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] YAN Ruqiang, SHEN Fei, SUN C, et al. Knowledge Transfer for Rotary Machine Fault Diagnosis[J]. IEEE Sensors Journal. 2020, 20(15): 8374-8393.
[2] LIANG Pengfei, DENG Chao, WU Jun, et al. Compound Fault Diagnosis of Gearboxes via Multi-label Convolutional Neural Network and Wavelet Transform[J]. Computers in Industry. 2019, 113: 103132.
[3] LIU Yunpeng, JIANG Hongkai, LIU Chaoqiang, et al. Data-augmented wavelet capsule generative adversarial network for rolling bearing fault diagnosis[J]. Knowledge-Based Systems. 2022, 252: 109439.
[4] 沈长青,王旭,王冬,等. 基于多尺度卷积类内迁移学习的列车轴承故障诊断[J]. 交通运输工程学报. 2020, 20(05): 151-164.
SHEN Changqing, WANG Xu, WANG Dong, et al. ZHU Zhong-kui. Multi-scale convolution intra-class transfer learning for train bearing fault diagnosis[J]. Journal of Traffic and Transportation Engineering. 2020, 20(5): 151-164.
[5] JIAN Chuanxia, YANG Kaijun, AO Yinhui. Industrial fault diagnosis based on active learning and semi-supervised learning using small training set[J]. Engineering Applications of Artificial Intelligence. 2021, 104: 104365.
[6] GUO Qun, ZHANG Xinhao, LI Jing, et al. Fault diagnosis of modular multilevel converter based on adaptive chirp mode decomposition and temporal convolutional network[J]. Engineering Applications of Artificial Intelligence. 2022, 107: 104544.
[7] Van M, Kang H. Bearing Defect Classification Based on Individual Wavelet Local Fisher Discriminant Analysis with Particle Swarm Optimization[J]. IEEE Transactions on Industrial Informatics. 2016, 12(1): 124-135.
[8] HE Qingbo, WU Enhao, PAN Yuanyuan. Multi-Scale Stochastic Resonance Spectrogram for fault diagnosis of rolling element bearings[J]. Journal of Sound and Vibration. 2018, 420: 174-184.
[9] 吴春志,江鹏程,冯辅周,等. 基于一维卷积神经网络的齿轮箱故障诊断[J]. 振动与冲击. 2018, 37(22): 51-56.
WU Chunzhi,JIANG Pengcheng,FENG fuzhou, et al. Faults diagnosis method for gearboxes based on a 1-D convolutional neural network[J]. Journal of Vibration and Shock. 2018,37(22):51-56.
[10] 张立智,徐卫晓,井陆阳,等. 基于二维深度卷积网络的旋转机械故障诊断[J]. 机械强度. 2020, 42(05): 1039-1044.
ZHANG Lizhi,XU Weixiao,JING Luyang, et al. Rotating machinery fault diagnosis based on two-demensional convolutional neural network[J]. Journal of Mechanical Strength. 2020, 42(05):1039-1044.
[11] HAN Tian, ZHANG Longwen, YIN Zhong, et al. Rolling bearing fault diagnosis with combined convolutional neural networks and support vector machine[J]. Measurement. 2021, 177: 109022.
[12] LI Xiang, ZHANG Wei, DING Qian. Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism[J]. Signal Processing. 2019, 161: 136-154.
[13] DING Yifei, JIA Minping, MIAO Qiuhua, et al. A novel time–frequency Transformer based on self–attention mechanism and its application in fault diagnosis of rolling bearings[J]. Mechanical Systems and Signal Processing. 2022, 168: 108616.
[14] QIAN Weiwei, LI Shunming. A novel class imbalance-robust network for bearing fault diagnosis utilizing raw vibration signals[J]. Measurement. 2020, 156: 107567.
[15] LI Guoqiang, WU Jun, DENG Chao, et al. Parallel multi-fusion convolutional neural networks based fault diagnosis of rotating machinery under noisy environments[J]. ISA Transactions. 2022, 128: 545-555.
[16] WANG Min, YU Hongtian, MIN Fan. Noise label learning through label confidence statistical inference[J]. Knowledge-Based Systems. 2021, 227: 107234.
[17] Singh S, Kumar N. Detection of Bearing Faults in Mechanical Systems Using Stator Current Monitoring[J]. IEEE Transactions on Industrial Informatics. 2017, 13(3): 1341-1349.
[18] PAN Haiyang, YANG Yu, LI Xin, et al. Symplectic geometry mode decomposition and its application to rotating machinery compound fault diagnosis[J]. Mechanical Systems and Signal Processing. 2019, 114: 189-211.
[19] LONG Yunyao, ZHOU Wuneng, LUO Yong. A fault diagnosis method based on one-dimensional data enhancement and convolutional neural network[J]. Measurement. 2021, 180: 109532.
[20] ZHANG Zhiqiang, YANG Qingyu, ZI Yanyang. Multi-scale and multi-pooling sparse filtering: A simple and effective representation learning method for intelligent fault diagnosis[J]. Neurocomputing. 2021, 451: 138-151.
[21] Dosovitskiy A, Beyer L, Kolesnikov A, et al. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale[J]. 2020.
[22] LIU Ze, LIN Yutong, CAO Yue, et al. Swin Transformer: Hierarchical Vision Transformer using Shifted Windows[J]. 2021.
[23] BAI Ruxue, XU Quansheng, MENG Zong, et al. Rolling bearing fault diagnosis based on multi-channel convolution neural network and multi-scale clipping fusion data augmentation[J]. Measurement. 2021, 184: 109885.
[24] XU Zifei, LI Chun, YANG Yang. Fault diagnosis of rolling bearings using an Improved Multi-Scale Convolutional Neural Network with Feature Attention mechanism[J]. ISA Transactions. 2021, 110: 379-393.
[25] Smith W A, Randall R B. Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark study[J]. Mechanical Systems and Signal Processing. 2015, 64-65: 100-131.
[26] FU Yang, ZHANG Yun, GAO Yuan, et al. Machining vibration states monitoring based on image representation using convolutional neural networks[J]. Engineering Applications of Artificial Intelligence. 2017, 65: 240-251.
[27] 卞景艺,刘秀丽,徐小力,等. 基于多尺度深度卷积神经网络的故障诊断方法[J]. 振动与冲击. 2021, 40(18): 204-211.
BIAN Jingyi, LIU Xiuli, XU Xiaoli,et al. Fault diagnosis method based on a multi-scale deep convolutional neural network.[J]. Journal of Vibration and Shock. 2021, 40(18): 204-211.
[28] 许子非,金江涛,李春. 基于多尺度卷积神经网络的滚动轴承故障诊断方法[J]. 振动与冲击. 2021, 40(18): 212-220.
XU Zifei,JIN Jiangtao,LI Chun. New method for the fault diagnosis of rolling bearings based on a multiscale convolutional neural network.[J]. Journal of Vibration and Shock. 2021, 40(18): 212-220.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}