陈其龙

• 1:广州商学院信息技术与工程学院

摘要(Abstract)：

滚动轴承故障的存在严重影响旋转类机械设备的运行安全，为了实现对滚动轴承故障的高效率诊断，这里引入了轻量级卷积神经网络SqueezeNet，用于识别滚动轴承的故障类型，并与传统的AlexNet及ResNet的性能指标进行了对比。利用凯斯西储大学的公开数据集作为神经网络的训练样本，分别对三种不同的神经网络进行了训练，以识别不同的滚动轴承故障类型。结果表明SqueezeNet在准确度上优于AlexNet，且训练效率是AlexNet的1.7倍，是ResNet的4.2倍，极大降低了网络的训练时间，并且SqueezeNet的本地模型所占的容量是AlexNet的6%，是ResNet的5%，大幅缩小了传统神经网络所需的存储空间。研究结果为实现神经网络方案在滚动轴承故障诊断中的移动端部署奠定了基础，同时为智能化机械设备的设计与制造提供了参考。

关键词(KeyWords)： 滚动轴承;故障诊断;卷积神经网络;轻量级;SqueezeNet

Abstract:

Keywords:

基金项目(Foundation): 广东省青年创新人才项目—基于深度学习的人体姿态识别判断技术研究（2020KQNCX105）;; 广东省普通高校特色创新项目—基于卷积神经网络的ROS机器人目标跟踪技术研究（2021KTSCX151）

作者(Author): 陈其龙

DOI: 10.19356/j.cnki.1001-3997.20260402.015

参考文献(References)：

• [1]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(7):134-143.(Qu Jianling,Yu Lu,Yuan Tao,et al.Adaptive fault diagnosis algorithm for rolling bearing based on one-dimensional convolutional neural network[J].Journal of Instrumentation,2018(7):134-143.)
• [2]董绍江，裴雪武，吴文亮，等.基于多层降噪技术及改进卷积神经网络的滚动轴承故障诊断方法[J].机械工程学报，2021,57(1):148-156.(Dong Shaojiang,Pei Xuewu,Wu Wenliang,et al.Fault diagnosis method of rolling bearing based on multi-layer noise reduction technology and improved convolutional neural network[J].Chinese Journal of Mechanical Engineering,2021,57(1):148-156.)
• [3]高统林，朱坚民，黄之文.基于卷积神经网络的滚动轴承故障诊断研究[J].农业装备与车辆工程，2019(12):115-120.(Gao Tonglin,Zhu Jianmin,Huang Zhiwen.Research on fault diagnosis of rolling bearing based on convolutional neural network[J].Agricultural Equipment and Vehicle Engineering,2019(12):115-120.)
• [4]Qu Jinxiu,Zhang Zhousuo,Gong Teng.A novel intelligent method for mechanical fault diagnosis based on dual-tree complex wavelet packet transform and multiple classifier fusion[J].Neurocomputing,2016,171(JAN.1):837-53.
• [5]Rafik Djemili a,Hocine Bourouba b,M.C.Amara Korba c,et al. Application of empirical mode decomposition and artificial neural network for automatic bearing faultdiagnosis based on vibration signals[J].Applied Acoustics,2015(89):16-27.
• [6]史庆军，郭晓振，刘德胜.基于特征量融合和支持向量机的轴承故障诊断[J].电子测量与仪器学报，2019(10):104-111.(Shi Qingjun,Guo Xiaozhen,Liu Desheng.Bearing fault diagnosis based on feature fusion and support vector machine[J].Journal of Electronic Measurement and Instrumentation,2019(10):104-111.)
• [7]阚涛，高哲，杨闯.采用分数阶动量的卷积神经网络随机梯度下降法[J].模式识别与人工智能，2020(6):559-567.(Kan Tao,Gao Zhe,Yang Chuang.Convolutional neural network stochastic gradient descent method using fractional momentum[J].Pattern Recognition and Artificial Intelligence,2020,33(6):9.)
• [8]宫文峰，陈辉，张泽辉，等.基于改进卷积神经网络的滚动轴承智能故障诊断研究[J].振动工程学报，2020(2):400-413.(Gong Wenfeng,Chen Hui,Zhang Zehui,et al.Research on intelligent fault diagnosis of rolling bearing based on improved convolutional neural network[J].Chinese Journal of Vibration Engineering,2020(2):400-413.)
• [9]昝涛，王辉，刘智豪，等.基于多输入层卷积神经网络的滚动轴承故障诊断模型[J].振动与冲击，2020(12):142-149.(Zan Tao,Wang Hui,Liu Zhihao,et al.Fault diagnosis model of rolling bearing based on multi-input layer convolutional neural network[J].Vibration and Shock,2020(12):142-149.)
• [10]PAN H,HE X,TANG S,et al. An improved bearing fault diagnosis method using one-dimensional CNN and LSTM[J].Journal of Mechanical Engineering.
• [11]张立智，井陆阳，徐卫晓，等.基于卷积降噪自编码器和CNN的滚动轴承故障诊断[J].组合机床与自动化加工技术，2019(6):58-62.(Zhang Lizhi,Jing Luyang,Xu Weixiao,et al.Fault diagnosis of rolling bearing based on convolutional noise reduction autoencoder and CNN[J]. Combined Machine Tool and Automated Machining Technology,2019(6):58-62.)
• [12]田科位，董绍江，姜保军，等.基于改进深度残差网络的轴承故障诊断方法[J].振动与冲击，2021(20):247-254.(Tian Kewei,Dong Shaojiang,Jiang Baojun,et al.Bearing fault diagnosis method based on improved deep residual network[J].Vibration and Shock,2021(20):247-254.)
• [13]王庆荣，杨磊，王松松.基于S变换和卷积神经网络的滚动轴承故障诊断[J].激光与光电子学进展，2021(22):65-74.(Wang Qingrong,Yang Lei,Wang Songsong.Fault diagnosis of rolling bearing based on S-transform and convolutional neural network[J].Advances in Lasers and Optoelectronics,2021(22):65-74.)
• [14]常亮，邓小明，周明全，等.图像理解中的卷积神经网络[J].自动化学报，2016(9):1300-1312.(Chang Liang,Deng Xiaoming,Zhou Mingquan,et al.Convolutional neural networks in image understanding[J]. Journal of Automation,2016(9):1300-1312.)
• [15]李坤伦，魏泽发，宋焕生.基于SqueezeNet卷积神经网络的车辆颜色识别[J].长安大学学报：自然科学版，2020(4):109-116.(Li Kunlun,Wei Zefa,Song Huansheng.Vehicle color recognition based on squee zenet convolutional neural network[J]. Journal of Chang’an University:Natural Science Edition,2020(4):109-116.)
• [16]Alex Krizhevsky,I Sutskever,G Hinton. ImageNet classification with deep convolutional neural networks[J].Advances in neural information processing systems,2012,25(2).
• [17]李晓峰，刘光中.人工神经网络BP算法的改进及其应用[J].四川大学学报：工程科学版，2000(2):105-109.(Li Xiaofeng,Liu Guangzhong.Improvement and application of artificial neural network BP algorithm[J].Journal of Sichuan University(Engineering Science Edition),2000(2):105-109.)
• [18]He Kaiming,Zhang Xiangyu,Ren Shaoqing,et al.Deep residual learning for image recognition;proceedings of the 2016 IEEE conference on computer vision and pattern recognition(CVPR),F,2016[C].
• [19]Forrest N.Iandola,Song Han,Matthew W.Moskewicz,et al.SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and<0.5MB model size[J].2016.
• [20]王继霄，李阳，王家宝，等.基于SqueezeNet的轻量级图像融合方法[J].计算机应用，2020(3):837-841.(Wang Jixiao,Li Yang,Wang Jiabao,et al.A lightweight image fusion method based on SqueezeNet[J]. Computer Applications,2020(3):837-841.)
• [21]Li Rui,Ran Chao,Zhang Bin,et al. Rolling bearings fault diagnosis based on improved complete ensemble empirical mode decomposition with adaptive noise,nonlinear entropy,and Ensemble SVM[J].Applied Sciences,2020,10(16):5542.