[1]张弘斌,袁奇,赵柄锡,等.采用多通道样本和深度卷积神经网络的轴承故障诊断方法[J].西安交通大学学报,2020,54(08):058-66.[doi:10.7652/xjtuxb202008008]
 ZHANG Hongbin,YUAN Qi,ZHAO Bingxi,et al.Bearing Fault Diagnosis with Multi-Channel Sample and Deep Convolutional Neural Network[J].Journal of Xi'an Jiaotong University,2020,54(08):058-66.[doi:10.7652/xjtuxb202008008]
点击复制

采用多通道样本和深度卷积神经网络的轴承故障诊断方法
分享到:

《西安交通大学学报》[ISSN:0253-987X/CN:61-1069/T]

卷:
54
期数:
2020年第08期
页码:
058-66
栏目:
出版日期:
2020-08-10

文章信息/Info

Title:
Bearing Fault Diagnosis with Multi-Channel Sample and Deep Convolutional Neural Network
文章编号:
0253-987X(2020)08-0058-09
作者:
张弘斌12 袁奇12 赵柄锡12 牛广硕12
1.西安交通大学能源与动力工程学院, 710049, 西安; 2.陕西省叶轮机械及动力装备工程实验室, 710049, 西安
Author(s):
ZHANG Hongbin12 YUAN Qi12 ZHAO Bingxi12 NIU Guangshuo12
1. School of Energy and Power Engineering, Xi'an Jiaotong University, Xi'an 710049, China; 2. Shaanxi Engineering Laboratory of Turbomachinery and Power Equipment, Xi'an Jiaotong University, Xi'an 710049, China
关键词:
轴承故障诊断 三通道样本 深度卷积神经网络 连续小波变换
Keywords:
bearing fault diagnosis three-channel samples deep convolutional neural network continuous wavelet transform
分类号:
TH17; TH133
DOI:
10.7652/xjtuxb202008008
文献标志码:
A
摘要:
提出了一种新的多通道样本构造方法,结合深度卷积神经网络来提高轴承故障诊断的效果。首先采用连续小波变换,分别提取了转子两端轴承振动信号的时频域特征,基于所得结果分别构造了针对两端轴承的单通道二维图形样本,并取上述两类单通道样本的均值构造了第3类单通道样本; 将得到的3类单通道样本融合,得到用于故障诊断的三通道样本; 建立不同结构的深度卷积神经网络,分别采用单通道样本和三通道样本对滚动轴承故障类型和故障严重程度进行诊断,并将所得结果进行对比分析。结果表明:在多种不同网络结构下,基于三通道样本的轴承故障诊断准确率均明显优于单通道样本,证明了提出的多通道样本构造方法在轴承故障诊断中有着更好的效果,可以为轴承故障诊断方法和样本构建提供参考。
Abstract:
A new multi-channel sample construction method combing with deep convolutional neural network is proposed to improve the effectiveness of bearing fault diagnosis. The time-frequency domain features of vibration signals from bearings at both ends of the rotor are extracted by continuous wavelet transform, and two types of signal-channel samples are constructed based on the obtained results. Then the third type of single-channel sample is produced by performing average operation on the two former types of single-channel samples. Furthermore, the obtained three types of single-channel samples are combined to obtain the three-channel sample adopted for fault diagnosis. The deep convolutional neural networks of different structures are established with the three types of single-channel samples and the three-channel samples respectively, the diagnosis is thus carried out on the rolling bearing with respect to different fault types and severity levels. A comparison between the diagnosis performances of single-channel samples and multi-channel samples indicates that the diagnosis accuracy of three-channel samples is obviously better than that of single-channel samples with different network structures. This approach provides a reference for bearing fault diagnosis in terms of sample construction.

参考文献/References:

[1] KANG M, KIM J, KIM J M. High-performance and energy-efficient fault diagnosis using effective envelope analysis and denoising on a general-purpose graphics processing unit [J]. IEEE Transactions on Power Electronics, 2015, 30(5): 2763-2776.
[2] MANJURUL ISLAM M M, KIM J M. Reliable multiple combined fault diagnosis of bearings using heterogeneous feature models and multiclass support vector Machines [J]. Reliability Engineering & System Safety, 2019, 184: 55-66.
[3] ZHAO Rui, WANG Dongzhe, YAN Ruqiang, et al. Machine health monitoring using local feature-based gated recurrent unit networks [J]. IEEE Transactions on Industrial Electronics, 2018, 65(2): 1539-1548.
[4] RAZAVI-FAR R, FARAJZADEH-ZANJANI M, SAIF M. An integrated class-imbalanced learning scheme for diagnosing bearing defects in induction motors [J]. IEEE Transactions on Industrial Informatics, 2017, 13(6): 2758-2769.
[5] SONG L Y, WANG H Q, CHEN P. Vibration-based intelligent fault diagnosis for roller bearings in low-speed rotating machinery [J]. IEEE Transactions on Instrumentation and Measurement, 2018, 67(8): 1887 -1899.
[6] KESKES H, BRAHAM A. Recursive undecimated wavelet packet transform and DAG SVM for induction motor diagnosis [J]. IEEE Transactions on Industrial Informatics, 2015, 11(5): 1059-1066.
[7] LI Yongbo, XU Mingqiang, LIANG Xihui, et al. Application of bandwidth EMD and adaptive multiscale morphology analysis for incipient fault diagnosis of rolling bearings [J]. IEEE Transactions on Industrial Electronics, 2017, 64(8): 6506-6517.
[8] TIAN Ye, MA Jian, LU Chen, et al. Rolling bearing fault diagnosis under variable conditions using LMD-SVD and extreme learning machine [J]. Mechanism and Machine Theory, 2015, 90: 175-186.
[9] 周小勇. 小波分析在故障诊断中的应用 [D]. 上海: 上海海运学院, 2001: 9-11.
[10] 顾杰. 基于小波分析的航空发动机振动故障诊断研究 [D]. 天津: 中国民航大学, 2016: 55-56.
[11] 周东翔, 张小明, 陈明, 等. 小波分析在发动机故障诊断中的应用 [J]. 内燃机与配件, 2019(21): 240-241.
ZHOU Dongxiang, ZHANG Xiaoming, CHEN Ming, et al. Application of wavelet analysis in engine fault diagnosis [J]. Internal Combustion Engine & Parts, 2019(21): 240-241.
[12] WANG Peng, Ananya, YAN Ruiqiang. Virtualization and deep recognition for system fault classification [J]. Journal of Manufacturing Systems, 2017, 44(S1): 310-316.
[13] ZHAO Rui, YAN Ruiqiang, CHEN Zhenghua, et al. Deep learning and its applications to machine health monitoring [J]. Mechanical Systems and Signal Processing, 2019, 115: 213-237.
[14] 王洋洋, 唐建, 代菊英, 等. 基于1DCNN的滚动轴承故障诊断方法 [J]. 机械管理开发, 2019, 34(10): 141-144, 159.
WANG Yangyang, TANG Jian, DAI Juying, at al. Fault diagnosis method of rolling bearing based on 1DCNN [J]. Mechanical Management and Development, 2019, 34(10): 141-144, 159.
[15] CHEN Zhiqiang, LI Chuan, SANCHEZ R V. Gearbox fault identification and classification with convolutional neural networks [J]. Shock and Vibration, 2015, 2015(2015): 1-10.
[16] 陈仁祥, 黄鑫, 杨黎霞, 等. 基于卷积神经网络和离散小波变换的滚动轴承故障诊断 [J]. 振动工程学报, 2018, 31(5): 883-891.
CHEN Renxiang, HUANG Xin, YANG Lixia, et al. Rolling bearing fault identification based on convolution neural network and discrete wavelet transform [J]. Journal of Vibration Engineering, 2018, 31(5): 883-891.
[17] WANG D, TSE P W, TSUI K L. An enhanced Kurtogram method for fault diagnosis of rolling element bearings [J]. Mechanical Systems and Signal Processing, 2013, 35(1/2): 176-199.
[18] 王斌. 基于小波分析和BP神经网络的滚动轴承故障诊断方法研究 [D]. 兰州: 兰州交通大学, 2014: 9-11.
[19] HUANG Ruyi, LIAO Yixiao, ZHANG Shaohui, et al. Deep decoupling convolutional neural network for intelligent compound fault diagnosis [J]. IEEE Access, 2019, 7: 1848-1858.
[20] QIU Guangqi, GU Yingkui, CAI Quan. A deep convolutional neural networks model for intelligent fault diagnosis of a gearbox under different operational conditions [J]. Measurement, 2019, 145: 94-107.
[21] 姚志英, 王成林, 姚滢滢. 基于深度卷积神经网络的物品检测研究 [J]. 物流技术, 2019, 38(8): 118-122.
YAO Zhiying, WANG Chenglin, YAO Yingying. Research on article inspection based on deep convolutional neural network [J]. Logistics Technology, 2019, 38(8): 118-122.
[22] Case Western Reserve University. Bearing data center, 2014, accessed: 2014-05-15 [EB/PL]. [2019-12-06]. http: ∥csegroups.case.edu/bearingdatacenter.
[23] AFRASIABI S, AFRASIABI M, PARANG B, et al. Real-time bearing fault diagnosis of induction motors with accelerated deep learning approach [C]∥2019 10th International Power Electronics, Drive Systems and Technologies Conference. Piscataway, NJ, USA: IEEE, 2019: 155-159.

相似文献/References:

[1]张西宁,向宙,唐春华.一种深度卷积自编码网络及其在滚动轴承故障诊断中的应用[J].西安交通大学学报,2018,52(07):001.[doi:10.7652/xjtuxb201807001]
 ZHANG Xining,XIANG Zhou,TANG Chunhua.A Deep Convolutional AutoEncoding Neural Network and Its Application in Bearing Fault Diagnosis[J].Journal of Xi'an Jiaotong University,2018,52(08):001.[doi:10.7652/xjtuxb201807001]
[2]张西宁,雷威,杨雨薇,等.采用自适应基因粒子群算法优化隐马尔科夫模型的方法及应用[J].西安交通大学学报,2018,52(08):001.[doi:10.7652/xjtuxb201808001]
 ZHANG Xining,LEI Wei,YANG Yuwei,et al.Adaptive Genetic Particle Swarm Algorithm for Optimization Hidden Markov Models with Applications[J].Journal of Xi'an Jiaotong University,2018,52(08):001.[doi:10.7652/xjtuxb201808001]

备注/Memo

备注/Memo:
收稿日期: 2020-01-12。作者简介: 张弘斌(1996—),男,硕士生; 袁奇(通信作者),男,教授,博士生导师。基金项目: 国家自然科学基金资助项目(11872289,11902237); 西安交通大学自然科学基金重点资助项目(ZRZD2017025)。
更新日期/Last Update: 2020-08-10