数模联合驱动的动态对抗自适应轴承故障诊断方法

PDF(2179 KB)

振动与冲击 ›› 2023, Vol. 42 ›› Issue (12) : 256-263.

论文

张锐奇，孙弋，于耀翔，郭亮，宗珠毓秀，高宏力

作者信息 +

A new simulation-data driven dynamic adversarial adaptive fault diagnosis method for bearings

ZHANG Ruiqi,SUN Yi,YU Yaoxiang,GUO Liang,ZONG Zhuyuxiu,GAO Hongli

Author information +

文章历史 +

摘要

针对现实工业场景下，故障数据样本稀缺，服役工况复杂导致的滚动轴承诊断准确率低下的问题，提出了一种数模联合驱动的动态对抗自适应轴承故障诊断方法。首先，提出用于快速产生具有明确时频域轴承故障特征的四自由度动力学仿真模型。随后，探讨了实测数据分布与动力学仿真信号之间的共性和差异性。最后，建立可以提取隐层域不变特征并自动对齐仿真源域数据、目标域待诊断数据分布的动态对抗自适应网络。设计滚动轴承故障诊断实验，探讨了由动力学模型产生带有标签信息的仿真信号、带标签信息的其他数据集实测信号与极少数带标签信息的待诊断实测信号构成的源域数据基础上，设定的三类任务中神经网络的诊断效果，完成了对大量无标签样本的分类识别。结果表明仿真信号包含轴承故障的特征信息，可以对真实的轴承数据进行表征，并且所提出的动态对抗自适应网络相较于其他诊断方法能更准确实现轴承的故障诊断。同时，源域数据中包含极少数的带标签目标域数据可使得提出方法的识别准确率大幅提升。

Abstract

A dynamic adversarial adaptive bearing fault diagnosis method driven by a combination of simulation and data is proposed to solve the problem of low diagnostic accuracy of rolling bearings due to the scarcity of fault data samples and complex service conditions in real industrial scenarios. Firstly, a four-degree-of-freedom dynamics simulation model was proposed for the rapid generation of bearing fault characteristics with a well-defined time-frequency domain. Subsequently, the commonalities and differences between the measured data distribution and the dynamics simulation signal were explored. Finally, a dynamic adversarial adaptive network that can extract invariant features in the hidden domain and automatically align the distribution of data in the source domain of the simulation and the data to be diagnosed in the target domain was developed. The fault diagnosis experiment of rolling bearing was designed, and the source domain data composed of the signal with label information generated by the dynamic model, the measured signal of other dataset with label information and a small amount of signal in target domain with label information. The diagnostic effect of the neural network in the three categories of tasks was discussed, and the classification and recognition of a large number of unlabeled samples were completed. The results show that the simulated signal contains the characteristic information of bearing faults, which can characterize the real bearing data, and the proposed dynamic confrontation adaptive network can realize the bearing fault diagnosis more accurately than other diagnosis methods. At the same time, the source domain data contains very few labeled target domain data, which can greatly improve the recognition accuracy of the proposed method.

导出引用

张锐奇，孙弋，于耀翔，郭亮，宗珠毓秀，高宏力. 数模联合驱动的动态对抗自适应轴承故障诊断方法[J]. 振动与冲击, 2023, 42(12): 256-263

ZHANG Ruiqi,SUN Yi,YU Yaoxiang,GUO Liang,ZONG Zhuyuxiu,GAO Hongli. A new simulation-data driven dynamic adversarial adaptive fault diagnosis method for bearings[J]. Journal of Vibration and Shock, 2023, 42(12): 256-263

参考文献

[1] 张建军，王仲生，芦玉华，等. 基于非线性动力学的滚动轴承故障工程建模与分析[J]. 振动与冲击，2010，29(11):30-34.
ZHANG Jian-jun, WANG Zhong-sheng, LU Yu-hua, et al. Nonlinear dynamic modeling for localized defects in a rolling element bearing[J]. Journal of Vibration and Shock, 2010, 29(11):30-34.
[2] ZHANG S, ZHANG S B, WANG B N, et al. Deep Learning Algorithms for Bearing Fault Diagnostics – A Comprehensive Review[J]. IEEE Access, 2020, 8: 29857-29881.
[3] LEI Y G, LI N P, GUO L, et al. Machinery health prognostics: A systematic review from data acquisition to RUL prediction[J]. Mechanical Systems and Signal Processing, 2018, 104:799-834.
[4] CHOW M Y, MANGUM P M. A neural network approach to real-time condition monitoring of induction motors[J]. IEEE Transactions on Industrial Electronics, 1991, 38(6):448-453.
[5] MALHI A, GAO R X. PCA-based feature selection scheme for machine defect classification[J]. IEEE Transactions on Instrumentation and Measurement, 2004, 53(6):1517-1525.
[6] HE D, LI R Y, ZHU J D, et al. Data Mining Based Full Ceramic Bearing Fault Diagnostic System Using AE Sensors[J]. IEEE Transactions on Neural Networks, 2011, 22(12):2022-2031.
[7] ROJAS A, NANDI A K. Detection and Classification of Rolling-Element Bearing Faults using Support Vector Machines[J] IEEE Workshop on Machine Learning for Signal Processing. IEEE, 2005:153-158
[8] GOYAL D, CHOUDHARY A, PABLA B S, et al. Support vector machines based non-contact fault diagnosis system for bearings[J]. Journal of Intelligent Manufacturing, 2020, 31(5):1275-1289
[9] 雷亚国. 基于改进Hilbert-Huang变换的机械故障诊断[J]. 机械工程学报，2011，47(5):7.
LEI Ya-guo. Machinery Fault Diagnosis Based on Improved Hilbert-Huang Transform[J]. Journal of Mechanical Engineering, 2011, 47(5):7.
[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] KRIZHEVSKY A, SUTSKEVER I, HINTON G. ImageNet Classification with Deep Convolutional Neural Networks[J]. Advances in neural information processing systems, 2012, 25(2).
[12] SZEGRDY C, LIU W, JIA Y Q, et al. Going Deeper with Convolutions [C]//2015 IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE,2015.
[13] SIMONYAN K, ZISSERMAN A. Very Deep Convolutional Networks for Large-Scale Image Recognition[J]. Computer Science, 2014.
[14] HE K M, ZHANG X Y, REN S Q, et al. Deep Residual Learning for Image Recognition[J]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016:770-778.
[15] HUANG G, LIU Z, LAURENS V D M, et al. Densely Connected Convolutional Networks[J]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2017:2261-2269.
[16] 曲建岭，余路，袁涛，等. 基于一维卷积神经网络的滚动轴承自适应故障诊断算法[J]. 仪器仪表学报，2018，39(7):134-143.
QU Jian-ling, YU Lu, YUAN Tao, et al. Adaptive fault diagnosis algorithm for rolling bearings based on one-dimensional convolutional neural network[J] Chinese Journal of Scientific Instrument, 2018, 39(7):134-143.
[17] 郭亮，高宏力，张一文，等. 基于深度学习理论的轴承状态识别研究[J]. 振动与冲击，2016，35(12): 166-170+195.
GUO Liang, GAO Hong-li, ZHANG Yi-wen, et al. Research on bearing condition monitoring based on deep learning[J]. Journal of Vibration and Shock, 2016, 35(12): 166-170+195.
[18] 郭亮，董勋，高宏力，等. 无标签数据下基于特征知识迁移的机械设备智能故障诊断[J]. 仪器仪表学报，2019，40(8):58-64.
GUO Liang, Dong Xun, Gao Hong-li, et al. Feature knowledge transfer based intelligent fault diagnosis method of machines with unlabeled data[J]. Chinese Journal of Scientific Instrument, 2019, 40(8):58-64.
[19] YU C H, WANG J D, CHEN Y Q, et al. Transfer Learning with Dynamic Adversarial Adaptation Network[J]. 2019 IEEE International Conference on Data Mining (ICDM), 2019: 778-786.
[20] GUO L, LEI Y G, XING S B, et al. Deep Convolutional Transfer Learning Network: A New Method for Intelligent Fault Diagnosis of Machines With Unlabeled Data[J]. IEEE Transactions on Industrial Electronics, 2019, 66(9):7316-7325.
[21] JIANG Y C, HUANG W T, LUO J N, et al. An improved dynamic model of defective bearings considering the three-dimensional geometric relationship between the rolling element and defect area[J]. Mechanical Systems and Signal Processing, 2019, 129:694-716
[22] LIU J, SHAO Y M, LIM T C. Vibration analysis of ball bearings with a localized defect applying piecewise response function[J]. Mechanism and Machine Theory, 2012, 56(1):156-169.
[23] 陈果. 转子-滚动轴承-机匣耦合系统中滚动轴承故障的动力学分析[J]. 振动工程学报，2008，21(6):11.
CHEN Guo. Dynamic analysis of ball bearing faults in rotor-ball bearing-stator coupling system[J]. Journal of Vibration Engineering, 2008, 21(6):11.
[24] 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: 100-131.
[25] QIU H, LEE J, LIN J, et al. Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics[J]. Journal of Sound and Vibration, 2006, 289(4-5):1066-1090.
[26] TZENG E, HOFFMAN J, ZHANG N, et al. Deep Domain Confusion: Maximizing for Domain Invariance[J]. ArXiv preprint ArXiv:1412.3474, 2014.
[27] GHIFARY M, KLEIJN W B, ZHANG M J. Domain Adaptive Neural Networks for Object Recognition[J]. PRICAI 2014: Trends in artificial intelligence: 13th Pacific Rim International Conference on Artificial Intelligence, 2014:898-904.
[28] GANIN Y, USTINOVA E, AJAKAN H, et al. Domain-Adversarial Training of Neural Networks[J]. Journal of Machine Learning Research, 2016, 17(59):1-35.