针对变工况条件下轴承故障数据无法大量获取以及诊断困难的问题,提出了基于变分模态分解和卷积神经网络的轴承故障诊断方法,使用稳态工况获取的数据训练,能对变工况下的数据实现有效诊断。首先对轴承振动信号进行变分模态分解,以获得有限带宽的固有模态函数;然后构建卷积神经网络模型,采用优化技术提高模型适应性,实现对固有模态函数的自适应特征提取和分类;最后使用台架试验获得的滚动轴承故障数据进行验证,并与深度残差网络和支持向量机进行对比。结果表明,该模型对变工况数据的诊断/识别率达到100%/98.86%,高于对比模型的测试结果,有效实现了变工况轴承故障诊断。
Abstract
To investigate the problem that it was difficult to obtain a large number of bearing fault data and diagnosefault type under varying operation conditions, a bearing fault diagnosis method based on variational mode decomposition and convolution neural network was proposed. This method could diagnose bearing data under varying operation conditions by using training data under steady conditions.Firstly, variational mode decomposition was used to decompose the bearing vibration signals in order to obtain a series of band-limited intrinsic modal functions.Then, convolution neural network was constructed to adaptiveextract and classifiy featuresof the IMFs, with optimization technology used to improve its adaptability.Finally, the rolling bearing fault data obtained from bench test was used in experimental verification, and the model of ResNet and SVM were used as comparison. The results showed that the diagnosis/recognition rate of the model is 100% / 98.86%under varying operation conditions that is higher than two comparison models, which also proved that the model can effectively realize bearing fault diagnosisunder varying operation conditions.
关键词
计量学 /
滚动轴承 /
复合型故障诊断 /
变工况 /
卷积神经网络 /
状态识别
Key words
metrology /
rolling bearing /
composite fault diagnosis /
varying operation conditions /
convolutional neural network /
state recognition
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
［1］Dragomiretskiy K, Zosso D. Variational Mode Decomposition［J］. IEEE Transactionson Signal Processing,2014, 62(3):531-544.
［2］张淑清,邢婷婷,何红梅,等.基于VMD及广义分形维数矩阵的滚动轴承故障诊断［J］.计量学报,2017,38(4):439-443.
Zhang S Q, Xing T T, He H M, et al. Bearing Fault Diagnosis Method Based on VMD and Generalized Fractal Dimension Matrix［J］. Acta Metrologica Sinica,2017,38(4):439-443.
［3］陈剑,庄学凯,吕伍佯,等.基于IVMD和马田系统的滚动轴承故障检测方法［J］.计量学报,2019,40(6):1083-1087.
Chen J, Zhuang X K, Lv W Y, et al. Fault Diagnosis of Rolling Bearing Using Mahalanobis-Taguchi System Based on IVMD［J］. Acta Metrologica Sinica,2019,40(6):1083-1087.
［4］Li F H, Li R, Tian L L, et al.Data-driven time-frequency analysis method based on variational mode decomposition and its application to gear fault diagnosis in variable working conditions［J］.Mechanical Systems and Signal Processing, 2019, 116:462-479.
［5］程淑红,张仕军,赵考鹏.基于卷积神经网络的生物式水质监测方法［J］.计量学报,2019,40(4):721-727.
Cheng S H, Zhang S J, Zhao K P. Biological Water Quality Monitoring Method Based on Convolution Neural Network［J］. Acta Metrologica Sinica, 2019,40(4):721-727.
［6］陈仁祥,黄鑫,杨黎霞,等.基于卷积神经网络和离散小波变换的滚动轴承故障诊断［J］.振动工程学报, 2018,31(5):883-891.
Chen R X, Huang X, Yang L X, et al. Rolling bearing fault identification based on convolution neural network and discrete wavelet transform［J］. Journal of Vibration Engineering, 2018, 31(5): 883-891.
［7］Liu C, Cheng G, Chen X H, et al. Planetary Gears Feature Extraction and Fault Diagnosis Method Based on VMD and CNN［J］. Sensors, 2018, 18(5): s18051522.
［8］雷亚国,杨彬,杜兆钧,等.大数据下机械装备故障的深度迁移诊断方法［J］.机械工程学报,2019,55(7): 1-8.
Lei Y G, Yang B, Du Z J, et al. Deep Transfer Diagnosis Method for Machinery in Big Data Era［J］. Journal of Mechanical Engineering,2019, 55(7): 1-8.
［9］胡晓依,荆云建,宋志坤,等.基于CNN-SVM的深度卷积神经网络轴承故障识别研究［J］.振动与冲击,2019,38(18):173-178.
Hu X Y, Jing Y J, Song Z K, et al. Bearing fault identification by using deep convolution neural networks based on CNN-SVM［J］. Journal of Vibration and Shock,2019, 38(18):173-178.
［10］Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting［J］. Journal of Machine Learning Research, 2014, 15(1):1929-1958.
［11］Nair V, Hinton G E. Rectified Linear Units Improve Restricted Boltzmann Machines［C］// Proceedings of the 27th International Conference on Machine Learning. Haifa: Omnipress, 2010: 807-814.
［12］Sandler M, Howard A, Zhu M, et al. MobileNetV2: Inverted Residuals and Linear Bottlenecks［C］// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2018: 4510-4520.
［13］陈剑,蔡坤奇,陶善勇,等.基于IITD模糊熵与随机森林的滚动轴承故障诊断方法［J］. 计量学报, 2021, 42(6): 774-779.
Chen J, Cai K q, Tao S Y, et al. Fault Diagnosis Method of Rolling Bearing Based on IITD Fuzzy Entropy and Random Forest［J］. Acta Metrologica Sinica, , 2021, 42(6): 774-779.
［14］He K M, Zhang X Y, Ren S Q, et al. Identity Mappings in Deep Residual Networks［C］// Proceedings of the European Conference on Computer Vision. Cham: Springer, 2016: 630-645.
基金
国家自然科学基金青年基金(11604070);安徽省重大科技项目(17030901049)
PDF(2531 KB)
