基于多尺度空洞卷积神经网络的滚动轴承故障识别方法

兰州理工大学学报 ›› 2025, Vol. 51 ›› Issue (3): 55-63.

基于多尺度空洞卷积神经网络的滚动轴承故障识别方法

汪小虎¹, 赵荣珍^*1,2, 邓林峰¹, 郑玉巧¹

1.兰州理工大学机电工程学院, 甘肃兰州 730050;
2.广州理工学院智能制造与电气工程学院, 广东广州 510540

收稿日期:2023-01-04 出版日期:2025-06-28 发布日期:2025-06-30
通讯作者: 赵荣珍(1960-),女,山东枣庄人,博士,教授,博导.Email:zhaorongzhen@lut.edu.cn
基金资助:
国家自然科学基金(52465012,62241308)

Fault recognition of rolling bearing based on multi scale atrous convolution-convolutional neural network

WANG Xiao-hu¹, ZHAO Rong-zhen^1,2, DENG Lin-feng¹, ZHENG Yu-qiao¹

1. School of Mechanical and Electronical Engineering, Lanzhou University of Technology, Lanzhou 730050, China;
2. College of Intelligent Manufacturing and Electronical Engineering, Guangzhou Institute of Science and Technology,Guangzhou 510540, China

Received:2023-01-04 Online:2025-06-28 Published:2025-06-30

摘要/Abstract

摘要： 针对现有卷积神经网络模型参数偏多导致滚动轴承智能诊断效率低和识别准确率受限于训练样本数量的问题,提出了基于多尺度空洞卷积神经网络的滚动轴承故障识别方法.该方法首先在模型的输入层采用大尺寸的空洞卷积核和标准卷积核提取一维振动信号的多尺度敏感特征,然后使用尺寸为1×1和3×1的小卷积核以及2×1的最大池化操作对输入层所提取敏感特征进一步提取深层抽象特征,最后用全局平均池化层代替传统卷积神经网络的全连接层.同时,分别采用西储大学轴承故障数据和实验室轴承故障数据进行实验验证.结果表明,该方法泛化性能良好,并且能够在训练样本较少的情况下出色地完成故障识别任务,即使在一定噪声干扰下也能够对轴承微弱故障准确识别.

关键词: 多尺度空洞卷积神经网络, 滚动轴承, 故障识别, 小样本, 微弱故障

Abstract: Aiming at the problems of low intelligent diagnosis efficiency of rolling bearings due to the excessive number of parameters of existing convolutional neural network models and the limited recognition accuracy due to the number of training samples, a rolling bearing fault recognition method based on multi-scale atrous convolution-convolutional neural network (MSAC-CNN) was proposed. In this method, a large atrous convolution kernel and a standard convolution kernel are used in the input layer of the model to extract the multi-scale sensitive features of one-dimensional vibration signals. Then, the sensitive features extracted in the input layer were further extracted by using the 1×1 and 3×1 small convolution kernel and the 2×1 maximum pooling operation. Finally, the fully connected layer in the traditional convolutional neural network is replaced by the global average pooling layer. The experimental verification results with bearing fault data from Western Reserve University and our laboratory show that the proposed method has good generalization performance and performs the task of fault identification with fewer training samples. Moreover, it can accurately identify weak bearing faults under certain noise interference.

Key words: multi scale atrous convolution-convolutional neural network, rolling bearing, fault identification, small samples, weak fault

中图分类号:

汪小虎, 赵荣珍, 邓林峰, 郑玉巧. 基于多尺度空洞卷积神经网络的滚动轴承故障识别方法[J]. 兰州理工大学学报, 2025, 51(3): 55-63.

WANG Xiao-hu, ZHAO Rong-zhen, DENG Lin-feng, ZHENG Yu-qiao. Fault recognition of rolling bearing based on multi scale atrous convolution-convolutional neural network[J]. Journal of Lanzhou University of Technology, 2025, 51(3): 55-63.

参考文献

[1] ZHANG J Q,SUN Y,GUO L,et al.A new bearing fault diagnosis method based on modified convolutional neural networks[J].Chinese Journal of Aeronautics,2020,33(2):439-447.
[2] LU Q D,YANG R,ZHONG M Y,et al.An improved fault diagnosis method of rotating machinery using sensitive features and RLS-BP neural network[J].IEEE Transactions on Instrumentation and Measurement,2020,69(4):1585-1593.
[3] XIA M,LI T,XU L,et al.Fault diagnosis for rotating machinery using multiple sensors and convolutional neural networks[J].IEEE/ASME Transactions on Mechatronics,2018,23(1):101-110.
[4] XIA M,LI T,SHU T X,et al.A two-stage approach for the remaining useful life prediction of bearings using deep neural networks[J].IEEE Transactions on Industrial Informatics,2019,15(6):3703-3711.
[5] 周冠禄,江永全,陈锦雄,等.基于卷积神经网络的故障诊断模型振动信号处理方法比较[J].中国科技论文,2020,15(7):729-734.
[6] 林天骄,宋浏阳,李石,等.基于空洞卷积神经网络的旋转机械故障诊断方法[J].测试技术,2020,39(12):98-104.
[7] 周翔宇,毛善君,李梅.基于频域降采样和CNN的轴承故障诊断方法[J].北京大学学报(自然科学版),2023,59(2):251-260.
[8] 崔灿,陈林波,张敏.CNN 混合模型的旋转机械故障诊断方法研究 [J].现代电子技术,2022,45(13):73-77.
[9] ALAWAD M,LIN M J.Stochastic-based deep convolutional networks with reconfigurable logic fabric[J].IEEE Transactions on Multi-Scale Computing Systems,2016,2(4):242-256.
[10] COMBES J M,GROSSMANN A,TCHAMITCHIAN P,et al.Wavelets:time-frequency methods and phase space[M].2nd ed.New York:Springer-Verlag,1989.
[11] CHEN L C,PAPANDREOU G,KOKKINOS I,et al.DeepLab: semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected CRFs[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848.
[12] PANG Y W,SUN M L,JIANG X H,et al.Convolution in convolution for network in network[J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(5):1587-1597.
[13] MAFI M,MARTIN H,CABRERIZO M,et al.A comprehensive survey on impulse and Gaussian denoising filters for digital images[J].Signal Processing,2019,157:236-260.
[14] LOPARO K A.Case Western Reserve University bearing data center seeded fault test data[DB/OL].[2020-07-11].https://engineering.case.edu/bearingdatacenter/download-data-file.
[15] ZHANG W,LI C H,PENG G L,et al.A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load[J].Mechanical Systems and Signal Processing,2018,100:439-453.
[16] ZHANG W,PENG G L,LI C H,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):425-445.
[17] 周奇才,刘星晨,赵炯,等.旋转机械一维深度卷积神经网络故障诊断研究[J].振动与冲击,2018,37(23):31-37.
[18] 卞景艺,刘秀丽,徐小力,等.基于多尺度深度卷积神经网络的故障诊断方法[J].振动与冲击,2021,40(18):204-211.
[19] 王琦,邓林峰,赵荣珍.基于改进一维卷积神经网络的滚动轴承故障识别[J].振动与冲击,2022,41(3):216-223.
[20] 谷然,陈捷,洪荣晶,等.基于改进自适应变分模态分解的滚动轴承微弱故障诊断[J].振动与冲击,2020,39(8):1-7.