受外界环境噪声以及振噪耦合的影响,滚动轴承早期故障信号特征微弱,对其实现智能故障诊断具有挑战性。为了解决上述问题,提出一种基于改进最大相关峭度解卷积(Improved maximum correlation kurtosis deconvolution, IMCKD)和多通道卷积神经网络(Multi-channel convolution neural network, MCCNN)的智能故障诊断方法。首先利用萤火虫算法并行搜寻最大相关峭度解卷积的两个影响参数,对原始振动信号进行自适应滤波,得到诊断用的数据源;然后将其输入到MCCNN中进行特征学习,不断更新网络参数;最后将特征应用于分类器识别,从而实现滚动轴承的智能故障诊断。为了验证方法的可行性和有效性,利用滚动轴承故障模拟试验台采集的数据对该算法进行了验证。试验结果表明,该方法能准确、有效地对滚动轴承的故障类型进行分类,即使在强背景噪声下仍具有90%以上的故障识别率,并具有较好的稳定性和泛化能力。
Abstract
Due to the influence of external environment noise and vibration noise coupling, the early faults vibration signals of the rolling bearing are very weak, which is a challenge to realize intelligent fault diagnosis. In order to solve the above problems, an intelligent fault diagnosis method based on the improved maximum correlation kurtosis deconvolution(IMCKD) and multi-channel convolution neural network(MCCNN) is proposed. Firstly, the firefly algorithm is used to search the two influence parameters of the maximum correlation kurtosis deconvolution in parallel, and the original vibration signals is adaptively filtered to get the diagnostic data source. Then it is input to MCCNN for feature learning so that the network parameters are continually updated. Finally, the features are applied to the classifier recognition to realize the intelligent fault diagnosis of rolling bearing. In order to verify the feasibility and effectiveness of the intelligent method, the data collected from the rolling bearing fault simulation test rig are used to verify the algorithm. The experimental results show that the method can classify the fault types of rolling bearing accurately and effectively, and it has more than 90% fault recognition rate even under strong background noise, and it has good stability and generalization ability.
关键词
滚动轴承 /
最大相关峭度解卷积 /
卷积神经网络 /
萤火虫优化 /
特征学习
{{custom_keyword}} /
Key words
Rolling bearing /
Maximum correlation kurtosis deconvolution /
Convolution neural network /
Firefly optimization /
Feature learning
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] CUI L, WANG X, XU Y, et al. A novel Switching Unscented Kalman Filter method for remaining useful life prediction of rolling bearing [J]. Measurement, 2019, 135:678-684.
[2] LEI Y, YANG B, JIANG X, et al. Applications of machine learning to machine fault diagnosis: A review and roadmap [J]. Mechanical Systems and Signal Processing, 2020,138: 106587.
[3] XU K, LI Y, LIU C, et al. Advanced Data Collection and Analysis in Data-Driven Manufacturing Process [J]. Chinese Journal of Mechanical Engineering, 2020,33(1):33-43.
[4] MCDONALD G L, ZHAO Q, ZUO M J. Maximum correlated Kurtosis deconvolution and application on gear tooth chip fault detection [J]. Mechanical Systems and Signal Processing, 2012, 33:237-255.
[5] LI Z, MING A, ZHANG W, et al. Fault Feature Extraction and Enhancement of Rolling Element Bearings Based on Maximum Correlated Kurtosis Deconvolution and Improved Empirical Wavelet Transform [J]. Applied Sciences, 2019, 9(9):1876.
[6] HUA W, LUO C, LENG J, et al. Mine gearbox fault diagnosis based on multiwavelets and maximum correlated kurtosis deconvolution[J]. Journal of Vibroengineering,2017,19(6):4185-4197.
[7] WANG F, LIU C, SU W, et al. Combined Failure Diagnosis of Slewing Bearings Based on MCKD-CEEMD-ApEn [J]. Shock and Vibration, 2018,2018:6321785.
[8] LECUN Y, BENGIO Y, HINTON G. Deep learning [J]. Nature, 2015, 521(7553):436-444.
[9] HINTON G, DENG L, YU D, et al. Deep Neural Networks for Acoustic Modeling in Speech Recognition [J]. IEEE Signal Processing Magazine, 2012,29(6): 82-97.
[10] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition [J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[11] INCE T, KIRANYAZ S, EREN L, et al. Real-Time Motor Fault Detection by 1-D Convolutional Neural Networks [J]. IEEE Transactions on Industrial Electronics, 2016,63(11):7067-7075.
[12] WANG D, GUO Q, SONG Y, et al. Application of Multiscale Learning Neural Network Based on CNN in Bearing Fault Diagnosis[J]. Journal of Signal Processing Systems,2019,91(10):1205-1217.
[13] 李恒,张氢,秦仙蓉等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J].振动与冲击,2018,37(19):124-131.
LI Heng, ZHANG Qing, QIN Xianrong, et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network [J]. Journal of Vibration and Shock, 2018,37(19):124-131.
[14] YANG XS. Firefly Algorithms for Multimodal Optimization [C]//Proceedings of the 5th International Symposium on Stochastic Algorithms-Foundations and Applications. Hokkaido Univ, Sapporo, Japan, 2009, 5792:169-178.
[15] NAIR V, HINTON G. Rectified linear units improve restricted boltzmann machines [C]//International Conference on Machine Learning(ICML). Haifa, Israel. Madison: Omnipress, 2010: 807-814.
[16] IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[J]. arXiv Preprint arXiv, 2015(1):1502.03167.
[17] ZHANG W, PENG G, LI C, et al. A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals [J]. Sensors, 2017, 17(2): 425.
[18] SIMONYAN K, ZISSERMAN A. Very Deep Convolutional Networks for Large-Scale Image Recognition [J]. arXiv Preprint arXiv, 2014,1:1409.1556.
[19] GU J, WANG Z, KUEN J, et al. Recent advances in convolutional neural networks [J]. Pattern Recognition, 2018,77:354-377.
[20] KINGMA D P, BA J. Adam: A method for stochastic optimization [J]. arXiv Preprint arXiv, 2014,1:1412.6980.
[21] JIANG G, HE H, YAN J, et al. Multiscale Convolutional Neural Networks for Fault Diagnosis of Wind Turbine Gearbox[J]. IEEE Transactions on Industrial Electronics, 2019,66(4):3196-3207.
[22] 崔玲丽,王鑫,王华庆等.基于改进开关卡尔曼滤波的轴承故障特征提取方法[J].机械工程学报,2019,55(7):44-51.
CUI Lingli, WANG Xin, WANG Huaqing, et al. Feature Extraction of Bearing Fault Based on Improved Switching Kalman Filter[J]. Journal of Mechanical Engineering, 2019,55(7):44-51.
[23] MAATEN L, HINTON G. Visualizing data using t-SNE [J]. Journal of Machine Learning Research, 2008,9: 2579-2605.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}