为解决滚动轴承保持架故障振动信号存在的不稳定性、无冲击特性和故障特征难以获取问题，研究提出基于“端到端”识别的适应性卷积神经网络故障诊断模型。将不同保持架故障状态下的振动信号按一定比例采用有重叠样本分割进行数据增强，并对样本实施分段标准化预处理构建训练和测试集合;利用卷积神经网络实现对振动信号的自适应特征提取和特征降维；在输出端利用全局平均池化替换经典构架中使用的全连接运算，以减少训练模型参数和过程运算量，避免发生过拟合，最终经Softmax分类输出诊断结果。试验结果表明算法能够达到99%以上的故障识别率，且在不同负载和转速下均保持良好的泛化性能和鲁棒性，可有效应用于轴承保持架故障诊断任务。
Abstract
Here, to solve existing problems of instability, non-impact characteristics and difficult to obtain fault characteristics in rolling bearing cage fault vibration signals, an adaptive convolution neural network fault diagnosis model based on "end-to-end" recognition was proposed. Firstly, vibration signals under different cage fault conditions were segmented by overlapping samples according to a certain proportion for data enhancement, and samples were preprocessed by subsection standardization to construct training and test sets. Then, the convolution neural network was used to realize adaptive feature extraction and feature dimensionality reduction of vibration signals. Finally, the global average pooling was used to replace the full connection operation used in the classical framework at the output end, reduce amounts of training model parameters and process calculation, and avoid over-fitting. The diagnosis results were output through Softmax classification. The test results showed that the algorithm can reach more than 99% fault recognition rate, maintain good generalization performance and robustness under different loads and rotating speeds, and be effectively applied in fault diagnosis of bearing cage.
关键词
保持架故障诊断 /
故障损伤程度 /
卷积神经网络 /
振动信号 /
故障诊断
{{custom_keyword}} /
Key words
cage fault diagnosis /
degree of fault damage /
convolution neural network /
vibration signal /
fault diagnosis
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
［1］中华人民共和国国务院. 国务院关于印发《中国制造2025》的通知［EB/OL］.(2015-05-08).http://www.gov.cn/zhengce/content/ 2015-05/19/ content_9784.htm.
［2］张伟.基于卷积神经网络的轴承故障诊断算法研究［D］.哈尔滨:哈尔滨工业大学,2017.
［3］汤芳,刘义伦,龙慧,等.基于改进小波包系数熵的保持架损伤程度识别［J］.计算机仿真,2018,35(2):360-365.
TANG Fang, LIU Yilun, LONG Hui, et al.Recognition of cage damage degree based on improved wavelet packet coefficient entropy［J］.Computer Simulation, 2018,35(2): 360-365.
［4］余路,曲建岭,高峰,等.基于改进稀疏编码的微弱振动信号特征提取算法［J］.仪器仪表学报,2017,38(3): 711-717.
YU Lu, QU Jianling, GAO Feng, et al. Weak vibration signal feature extraction algorithm based on improved sparse coding［J］.Chinese Journal of Scientific Instrument, 2017,38(3): 711-717.
［5］李舜酩,郭海东,李殿荣.振动信号处理方法综述［J］.仪器仪表学报,2013,34(8):1907-1915.
LI Shunming, GUO Haidong, LI Dianrong.Summary of vibration signal processing methods［J］.Chinese Journal of Scientific Instrument, 2013,34(8): 1907-1915.
［6］胡晓扬.基于时频特征的旋转装备故障诊断技术研究［D］.哈尔滨:哈尔滨工业大学,2019.
［7］INOUE T, SUEOKA A, KANEMOTO H, et al. Detection of minute signs of a small fault in a periodic or a quasi-periodic signal by the harmonic wavelet transform［J］. Mechanical Systems and Signal Processing，2007，21(5):2041-2055.
［8］钟先友.旋转机械故障诊断的时频分析方法及其应用研究［D］.武汉:武汉科技大学,2014.
［9］陈佩.主成分分析法研究及其在特征提取中的应用［D］.西安:陕西师范大学,2014.
［10］陈保家,朱晨希,刘浩涛,等.基于LLTSA算法维数约简的滚动轴承故障诊断［J］.机械设计与研究,2017,33(5):110-114.
CHEN Baojia, ZHU Chenxi, LIU Haotao, et al. Fault diagnosis of rolling bearing based on dimension reduction of LLTSA algorithm［J］. Mechanical Design and Research, 2017, 33(5): 110-114.
［11］王雷.基于流形学习的滚动轴承故障诊断若干方法研究［D］.大连:大连理工大学,2013.
［12］丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述［J］.电子科技大学学报,2011,40(1):2-10.
DING Shifei, QI Bingjuan, TAN Hongyan. A review of support vector machine theory and algorithms［J］.Journal of University of Electronic Science and Technology of China, 2011,40(1): 2-10.
［13］桑应宾. 基于K近邻的分类算法研究［D］.重庆:重庆大学,2009.
［14］胡金滨,唐旭清.人工神经网络的BP算法及其应用［J］.信息技术,2004,28(4):1-4.
HU Jinbin, TANG Xuqing.BP algorithm of artificial neural network and its application ［J］ .Information Technology, 2004,28(4): 1-4.
［15］李辉,刘义伦,唐德尧.基于冲击特征的轴承保持架变形故障识别方法研究［J］.机车电传动,2015(6):95-97.
LI Hui, LIU Yilun, TANG Deyao.Research on identification method of bearing cage deformation failure based on impact characteristics［J］. Electric Drive for Locomotives, 2015(6): 95-97.
［16］KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classifification with deep convolutional neural networks［C］//26th Annual Conference on Neural Information Processing Systems. Lake Tahoe: NIPS, 2012.
［17］PALAZ D, MAGIMAI M D, COLLOBERT R. Convolutional neural networks-based continuous speech recognition using raw speech signal［C］//40th IEEE International Conference on Acoustics, Speech and Signal. South Brisbane: IEEE, 2015.
［18］CHEN Z Q， LI C， SANCHEZ R． Gearbox fault identification and classification with convolutional neural networks［J］． Shock and Vibration，2015(2): 1-10．
［19］袁建虎,韩涛,唐建,等.基于小波时频图和CNN的滚动轴承智能故障诊断方法［J］.机械设计与研究,2017, 33(2):93-97.
YUAN Jianhu, HAN Tao, TANG Jian, et al. Intelligent fault diagnosis method of rolling bearing based on wavelet time-frequency graph and CNN［J］. Mechanical Design and Research, 2017, 33(2): 93-97.
［20］杨观赐,杨静,李少波,等.基于Dopout与ADAM优化器的改进CNN算法［J］.华中科技大学学报(自然科学版), 2018,46(7):122-127.
YANG Guanci, YANG Jing, LI Shaobo, et al. Improved CNN algorithm based on Dopout and ADAM optimizer［J］.Journal of Huazhong University of Science and Technology (Natural Science Edition), 2018,46(7): 122-127.
［21］周飞燕,金林鹏,董军.卷积神经网络研究综述［J］.计算机学报,2017,40(6):1229-1251.
ZHOU Feiyan, JIN Linpeng, DONG Jun. A review of research on convolutional neural networks［J］. Chinese Journal of Computers, 2017,40(6): 1229-1251.
［22］LU C，WANG Z Y，QIN W L，et al． Fault diagnosis of rotary machinery components using a stacked denoising autoencoder-based health state identification［J］．Signal Processing，2017，130: 377-388．
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(3497 KB)
