深度卷积神经网络在滑动轴承转子轴心轨迹识别中的应用

PDF(1895 KB)

振动与冲击 ›› 2021, Vol. 40 ›› Issue (3) : 233-239.

论文

郭明军1,李伟光1,杨期江2,赵学智1

作者信息 +

Application of deep convolution neural network in identification of journal bearing rotor center orbit

GUO Mingjun1, LI Weiguang1, YANG Qijiang2, ZHAO Xuezhi1

Author information +

文章历史 +

摘要

针对传统旋转机械智能识别方法需要人为提取特征及诊断精度低的问题，基于深度学习的强大学习能力，提出一种深度卷积神经网络故障诊断模型（deep convolutional neural network fault diagnosis model，DCNN-FDM）用于轴心轨迹识别。该模型包括输入模块、特征提取模块及分类模块三部分。原始图像输入模型后，经过输入模块的二值化处理及最近邻插值，统一变为尺寸大小为32x32的单通道图像；然后，经特征提取模块中两组交替的卷积层和池化层作用，得到图形特征；最后，这些特征经全连接层的扁平化处理而张成一维向量，输入到softmax分类器中进行分类。利用奇异值差分谱方法，对实测轴心轨迹进行提纯，得到4类轴心轨迹样本集用于DCNN-FDM的训练与预测。结果表明：本文所提模型较传统的浅层学习模型的识别效果好，可实现转子故障的精确诊断，识别率达到97.09%。最后通过全连接层的主成分可视化分析，验证了模型具备自适应特征学习能力。

Abstract

Here, aiming at the problem of traditional rotating machinery intelligent identification method requiring artificial feature extraction with low diagnostic accuracy, based on deep learning having strong learning ability, a deep convolutional neural network fault diagnosis model (DCNN-FDM) was proposed to identify rotor axis center orbit. The model included 3 parts of input module, feature extraction one and classification one. After the original images were input into the model to do binary processing and nearest neighbor interpolation of input module, they were unified into a single channel images with size of 32×32. Then, through interaction of two groups of alternating convolution layers and pooling layers in feature extraction module, the graphic features were obtained. Finally, through flattening treatment of the full connection layer, these features were expanded into one-dimensional vectors to be input into SOFTMAX classifier for classification. By using the singular value difference spectrum method, four kinds of axis center orbit samples were obtained for training and prediction of DCNN-FDM. The results showed that the recognition effect of the proposed model is better than that of the traditional shallow learning model to realize the accurate diagnosis of rotor faults, and the recognition rate reaches 97.09%; the visual analysis of main components of the full connection layer verifies the model having the adaptive feature learning ability.

导出引用

郭明军1,李伟光1,杨期江2,赵学智1. 深度卷积神经网络在滑动轴承转子轴心轨迹识别中的应用[J]. 振动与冲击, 2021, 40(3): 233-239

GUO Mingjun1, LI Weiguang1, YANG Qijiang2, ZHAO Xuezhi1. Application of deep convolution neural network in identification of journal bearing rotor center orbit[J]. Journal of Vibration and Shock, 2021, 40(3): 233-239

参考文献

[1] 杨期江,李伟光,郑相立,王凯.变支点滑动轴承工作机理分析及转子振动特性试验研究[J]. 华南理工大学学报(自然科学版),2016,44(11):103-112.
YANG Qijiang, LI Weiguang, ZHENG Xiangli, et al. Analysis of work Mechanism of variable pivot Journal bearing and experimental investigation into vibration characteristic of rotor [J]. Journal of South China University of Technology (Natural Science Edition), ,2016,44(11):103-112.
[2] 赵伟. 油膜支承可倾瓦轴承—转子系统动力学特性分析[D].华南理工大学,2016.
ZHAO Wei. Dynamic behaviors analysis of oil film-pivot tilting-pad bearing-rotor system [D]. South China University of Technology,2016
vibration and shock, 2019,38(4):199-205.
[3] 袁倩,孙冬梅,范文.基于D-S证据理论的轴心轨迹自动识别方法[J].机床与液压,2017,45(07):167-171.
YUAN Qian, SUN Dongmei, FAN Wen. Automatic identification method of axis orbits based on D-S evidential theory [J]. Machine tool & Hydraulic, 2017,45(07):167-171.
[4] 李辉,白亮,罗兴錡,贾嵘,田录林.基于模糊聚类的水电机组轴心轨迹多重分形特征识别[J].水力发电学报,2012,31(04):238-242.
LI Haui, BAI Liang, LUO Xingqi, et al. Multi-fractal feature recognition for shaft centerline orbit of hydropower units based on fuzzy clustering [J]. Journal of Hydroelectric Engineering, 2012,31(04):238-242.
[5] 李友平,陈启卷.基于灰色理论与不变性矩的水电机组轴心轨迹自动识别[J].电力系统自动化,2001, 25(09):19-22.
LI Youping, CHEN Qijuan. Automatic identification of axis orbit of hydroelectric generating set based on grey theory and moment invariants [J]. Journal of Automation of Electric Power Systems , 2001,25(09):19-22.
[6]万鹏,王红军,徐小力.局部切空间排列和支持向量机的故障诊断模型[J].仪器仪表学报,2012,33(12):2789-2795.
WAN Peng, WANG Hongjun, XU Xiaoli. Fault diagnosis model based on local tangent space alignment and support vector machine [J]. Chinese Journal of Scientific Instrument, 2012,33(12):2789-2795.
[7] 郭鹏程,罗兴锜,王勇劲,白亮,李辉.基于粒子群算法与改进BP神经网络的水电机组轴心轨迹识别[J].中国电机工程学报,2011,31(08):93-97.
GUO Pengcheng, LUO Xingqi, WANG Yongjin, et al. Identification of shaft centerline orbit for hydropower units based on particle swarm optimization and improved BP neural network [J]. Proceedings of The Chinese Society for Electrical Engineering, 2011,31(08):93-97.
[8] 李松柏,康子剑,陶洁.基于信息融合及堆栈降噪自编码的齿轮故障诊断[J].振动与冲击,2019,38(05):216-221.
LI Xiong, TANG Zijian, TAO Jie. Gear fault diagnosis based on information fusion and stacked de-noising auto- encoder [J]. Journal of vibration and shock, 2019, 38(05):216-221.
[9] HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. science, 2006, 313(5786): 504-507.
[10] 李巍华,单外平,曾雪琼.基于深度信念网络的轴承故障分类识别[J].振动工程学报,2016,29(02):340-347.
LI Weihua, SHAN Waiping, ZENG Xueqiong. Bearing fault identification based on deep belief network [J]. Chinese Journal of Vibration Engineering, 2016,29(02):340-347.
[11] 刘星辰,周奇才,赵炯,沈鹤鸿,熊肖磊.一维卷积神经网络实时抗噪故障诊断算法[J].哈尔滨工业大学学报,2019,51(07):89-95.
LIU Xingchen, ZHOU Qicai, ZHAO Jiong, et al. Real-time and anti-noise fault diagnosis algorithm based on 1-D convolutional neural network [J]. Journal of Harbin Institute of Technology, 2019,51(07):89-95.
[12] 曲建岭,余路,袁涛,田沿平,高峰.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(07):134-143.
QU Jiangling, 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(07):134-143.
[13] CHEN Z Q, LI C, SANCHEZ R V. Gearbox fault identification and classification with convolutional neural networks[ J]. Shock and Vibration, 2015, 2015(1), 1:11.
[14] 赵学智，叶邦彦，陈统坚. 奇异值差分谱理论及其在车床主轴箱故障诊断中的应用[J]. 机械工程学报，2010,46(1):100-108.
ZHAO Xue-zhi, YE Bang-yan, CHEN Tong-jian. Difference spectrum theory of singular value and its application to the fault diagnosis of headstock of lathe [J]. Journal of Mechanical Engineering, 2010, 46(1):100-108.
[15] 张景润，李伟光，李振，赵学智. 基于奇异值差分谱理论的大型转子轴心轨迹提纯[J].振动与冲击，2019，38(4):199-205.
ZHANG Jing-run, LI Wei-guang, LI Zhen, et al. Purification a large rotor axis's based on the difference spectrum theory of singular value [J]. Journal of vibration and shock, 2019,38(4):199-205.