基于残差连接和1D-CNN的滚动轴承故障诊断研究

PDF(1287 KB)

振动与冲击 ›› 2021, Vol. 40 ›› Issue (10) : 1-6.

论文

赵敬娇1，赵志宏1,2，杨绍普1,2

作者信息 +

Rolling bearing fault diagnosis based on residual connection and 1D-CNN

ZHAO Jingjiao1，ZHAO Zhihong1,2，YANG Shaopu1,2

Author information +

文章历史 +

摘要

针对滚动轴承故障诊断人工提取特征困难、浅层诊断模型性能差的问题，提出一种基于残差连接的一维卷积神经网络(1D-CNN)的深层轴承故障诊断模型。将原始轴承振动信号输入网络中，利用具有残差连接的多个一维卷积层自动提取特征，残差连接能够在提取深层特征信息的同时将浅层提取的特征信息保留下来，与无残差连接的一维卷积网络相比能获得更丰富的轴承信号特征信息，并输入到Softmax层进行分类，输出轴承振动信号的故障类型。该研究通过不同残差网络结构模型的设计，验证具有残差连接的1D-CNN的网络模型在轴承故障诊断的有效性。试验结果表明，残差连接能有效提高轴承故障诊断的准确率。

Abstract

Aiming at the difficulty of manual feature extraction and the poor performance of shallow fault diagnosis models, a deep fault diagnosis model based on one-dimensional convolutional neural network (1D-CNN) and residual connection was proposed.A bearing vibration signal was input into the network, then the feature can be automatically extracted by one-dimensional convolution layers with residual connection.The residual connection can not only extract deep features and input it into the deep network layer, but also preserve the features extracted from the shallow layer.Compared with the 1D-CNN network without residual connection, it can obtain more abundant feature information.A Softmax layer was designed to classify the fault type of the vibration signal and output the result.By designing different diagnosis models, the validity of the 1D-CNN network with residual connection in the bearing fault diagnosis was verified, and finally it is proved that the residual connection can effectively improve the fault diagnosis results.

导出引用

赵敬娇1，赵志宏1,2，杨绍普1,2. 基于残差连接和1D-CNN的滚动轴承故障诊断研究[J]. 振动与冲击, 2021, 40(10): 1-6

ZHAO Jingjiao1，ZHAO Zhihong1,2，YANG Shaopu1,2. Rolling bearing fault diagnosis based on residual connection and 1D-CNN[J]. Journal of Vibration and Shock, 2021, 40(10): 1-6

参考文献

［1］刘亚.数据驱动的滚动轴承故障诊断与健康状态评估［D］.济南：山东大学，2019.

［2］RAI A, UPADHYAY S H.A review on signal processing techniques utilized in the fault diagnosis of rolling element bearings［J］.Tribology International, 2016,96: 289-306.

［3］梅宏斌.滚动轴承振动监测与诊断［M］.北京:机械工业出版社，1995.

［4］俞培松.滚动轴承振动故障诊断技术的研究及其实际应用［D］.上海：同济大学，2007.

［5］TAMILSELVAN P, WANG P F.Failure diagnosis using deep belief learning based health state classification［J］.Reliability Engineering & System Safety, 2013,115: 124-135.

［6］杨宇，于德介，程军圣.基于EMD与神经网络的滚动轴承故障诊断方法［J］.振动与冲击，2005,24(1): 85-88.
YANG Yu, YU Dejie, CHENG Junsheng.Roller bearing fault diagnosis method based on EMD and neural network［J］.Journal of Vibration and Shock, 2005,24(1): 85-88.

［7］杜小磊，陈志刚，许旭，等.基于小波卷积自编码器和LSTM网络的轴承故障诊断研究［J］.机电工程，2019,36(7): 663-668.
DU Xiaolei, CHEN Zhigang, XU Xu, et al.Fault diagnosis of bearing based on wavelet convolutional auto-encoder and LSTM network［J］.Journal of Mechanical & Electrical Engineering, 2019,36(7): 663-668.

［8］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.

［9］李恒，张氢，秦仙蓉，等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法［J］.振动与冲击，2018,37(19): 124-131.
LI Heng, ZHANG Qing, QIN Xianrong, et al.Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network［J］.Journal of Vibration and Shock, 2018,37(19): 124-131.

［10］PAN H H, HE X X, TANG S, et al.An improved bearing fault diagnosis method using one-dimensional CNN and LSTM［J］.Journal of Mechanical Engineering, 2018,64(7/8): 443-452.

［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.Fault diagnosis of rolling bearing based on convolutional denoising auto-encoder and CNN［J］.Modular Machine Tool & Automatic Manufacturing Technique, 2019(6): 58-62.

［12］赵志成，罗泽，王鹏，等.基于深度残差网络图像分类算法研究综述［J］.计算机系统应用，2020,29(1): 14-21.
ZHAO Zhicheng, LUO Ze, WANG Peng, et al.Survey on image classification algorithms based on deep residual network［J］.Computer Systems & Applications, 2020,29(1): 14-21.

［13］安晶，艾萍，徐森，等.一种基于一维卷积神经网络的旋转机械智能故障诊断方法［J］.南京大学学报(自然科学)， 2019,55(1): 133-142.
AN Jing, AI Ping, XU Sen, et al.An intelligent fault diagnosis method for rotating machinery based on one dimensional convolution neural network［J］.Journal of Nanjing University (Natural Science), 2019,55(1): 133-142.

［14］赵璐，马野.基于一维卷积神经网络的齿轮箱故障诊断研究［J］.测试技术学报，2019,33(4): 302-306.
ZHAO Lu, MA Ye.Fault diagnosis of gear box based on one-dimensional convolutional neural networks［J］.Journal of Test and Measurement Technology, 2019,33(4): 302-306.

［15］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 Jianling, YU Lu, YUAN Tao.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.

［16］陈伟.深度学习在滚动轴承故障诊断中的应用研究［D］.成都：西南交通大学，2018.

［17］HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition［C］//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas: IEEE, 2016.

［18］刘建伟,赵会丹,罗雄麟,等.深度学习批归一化及其相关算法研究进展［J］.自动化学报, 2020,46(6): 1090-1120.
LIU Jianwei, ZHAO Huidan, LUO Xionglin, et al.Research progress on batch normalization of deep learning and its related algorithms［J］.Acta Automatica Sinica, 2020,46(6): 1090-1120.

［19］BJORCK N, GOMES C, SELMAN B, et al.Understanding batch normalization［C］//32nd Conference on Neural Information Processing Systems.Montréal: NeurIPS, 2018.

［20］HINTON G E, SRIVASTAVA N, KRIZHEVSKY A, et al.Improving neural networks by preventing co-adaptation of feature detectors［J］.Computer Science, 2012,3(4): 212-223.

［21］MAATEN L, HINTON G.Visualizing data using t-SNE［J］.Journal of Machine Learning Research, 2008,9: 2579-2605.