Journal of Shandong University(Engineering Science) ›› 2019, Vol. 49 ›› Issue (2): 80-87.doi: 10.6040/j.issn.1672-3961.0.2018.268

• Machine Learning & Data Mining • Previous Articles     Next Articles

Research onfeature selection technology in bearing fault diagnosis

Jiachen WANG1(),Xianghong TANG1,2,3,*(),Jianguang LU1,2,3   

  1. 1. Key Laboratory of Advanced Manufacturing Technology, Ministry of Education, Guizhou University, Guiyang 550025, Guizhou, China
    2. School of Mechanical Engineering, Guizhou University, Guiyang 550025, Guizhou China
    3. State Key Laboratory of Public Big Data, Guizhou University, Guiyang 550025, Guizhou China
  • Received:2018-08-03 Online:2019-04-20 Published:2019-04-19
  • Contact: Xianghong TANG E-mail:wangjiachen512@qq.com;xhtang@gzu.edu.cn
  • Supported by:
    贵州省公共大数据重点实验室开放基金资助项目(2017BDKFJJ019);贵州大学引进人才基金资助项目(贵大人基合字(2016)13号)

RichHTML

102

PDF (PC)

510

Abstract:

A new method based on feature selection (FS) was proposed to select efficient features to promote the classification accuracy in bearing fault diagnosis. First, the outstanding features whose classification accuracy were higher than the threshold were directly selected by diagnosis model from a big feature set. Then the significant combinations of features which had less dimensions and higher classification accuracy were selected in the candidate feature set by a distinctive feature-oriented manner. Experiments showed that the proposed method had advantages in selecting efficient features, reducing the model parameters, decreasing the demand of samples and enhancing the model classification accuracy. As a result, it provided a new idea for feature selection and improved the efficiency of bearing fault diagnosis.

Key words: rolling bearing, fault diagnosis, outstanding features, outstanding features combination, feature selection

CLC Number: 

  • TP391.4

Fig.1

Diagram of outstanding features combination"

Fig.2

Experimental test rig of CWRU"

Table 1

Samples dataset of CWRU"

轴承状态 样本个数 样本标签
正常 150 0
外圈故障 150 1
内圈故障 150 2
球体故障 150 3

Fig.3

Experimental Test Rig of CUT-2"

Fig.4

Location of 6900ZZ bearing faults"

Table 2

Samples dataset of CUT-2"

轴承状态 样本个数 样本标签
正常 200 0
外圈故障 200 1
内圈故障 200 2
球体故障 200 3

Fig.5

Bearingfault diagnosis method based on outstanding features combination"

Table 3

Set of outstanding features parameters"

诊断阶段 特征种类个数 显著特征(组合) 组合个数 人工设定准确率阈值
1 1 K、V、S、E 4 0.5
2 2 K_V、M_V、V_R、V_Ma、V_Pe、V_Pu、V_E、V_S 26 0.8
3 3 M_K_V、K_V_S、K_V_R、K_V_Ma、K_V_Pe、K_V_Pu、K_V_E、M_V_Pu、V_R_Pe、V_S_E 28 0.9
4 6 M_K_V_R、K_V_S_Ma、K_V_R_Ma、K_V_R_Pe、K_V_R_Pu、K_V_Ma_Pe(其中K_V_S_Ma和K_V_Ma_Pe准确度相同) 36 前五位
5 5 M_K_V_R_Ma、M_K_V_R_Pe、K_V_R_Ma_Pe、K_V_R_Ma_Pu 21 0.95
6 6 M_K_V_R_Ma_Pe、M_K_V_R_Ma_Pu、K_V_R_Ma_Pe_Pu 12 0.955
7 7 M_K_V_R_Ma_Pe_Pu、M_K_V_S_R_Ma_Pu、K_V_S_R_Ma_Pe_Pu 7 前三位
8 8 M_K_V_S_R_Ma_Pe_Pu
M_K_V_R_Ma_Pe_Pu_E
M_K_V_S_R_Ma_Pu_E
K_V_S_R_Ma_Pe_Pu_E		 4 前三位
9 9 M_K_V_S_R_Ma_Pe_Pu_E 诊断结束，无参数设置 诊断结束，无参数设置

Table 4

Result ofsingle feature diagnosis"

特征种类 平均分类准确率/% 是否显著特征
Mean 42.62 否
Kurtosis 63.12 是
Var 87.62 是
Skewness 57.16 是
RMS 42.12 否
Margin 41.37 否
Peak 33.37 否
Pulse 48.33 否
Edge 65.66 是

Table 5

Result oftwo types features combination diagnosis"

特征种类 平均分类准确率/% 是否显著特征组合
M_K 71.37 否
K_V 97.37 是
K_S 74.20 否
K_R 64.00 否
K_Ma 66.50 否
K_Pe 62.70 否
K_Pu 71.41 否
K_E 75.95 否
M_V 83.95 是
V_R 87.50 是
V_Ma 85.54 是
V_Pe 86.87 是
V_Pu 83.25 是
V_E 81.58 是
M_S 62.50 否
V_S 89.25 是
S_R 62.00 否
S_Ma 56.87 否
S_Pe 57.16 否
S_Pu 58.75 否
S_E 68.79 否
M_E 65.62 否
R_E 67.70 否
Ma_E 66.62 否
Pe_E 65.08 否
Pu_E 72.04 否

Table 6

Result ofthree types features combination diagnosis"

特征种类 平均分类准确率/% 是否显著特征组合
M_K_V 96.54 是
K_V_S 96.62 是
K_V_R 96.95 是
K_V_Ma 96.33 是
K_V_Pe 96.12 是
K_V_Pu 95.79 是
K_V_E 94.66 是
M_V_S 85.54 否
M_V_R 85.29 否
M_V_Ma 84.29 否
M_V_Pe 84.83 否
M_V_Pu 90.50 是
M_V_E 81.04 否
V_S_R 88.25 否
V_R_Ma 84.54 否
V_R_Pe 92.00 是
V_R_Pu 85.29 否
V_R_E 82.12 否
V_S_Ma 86.33 否
V_Ma_Pe 84.12 否
V_Ma_Pu 85.66 否
V_Ma_E 81.33 否
V_S_Pe 87.70 否
V_Pe_Pu 84.41 否
V_Pe_E 82.54 否
V_S_Pu 86.50 否
V_Pu_E 83.12 否
V_S_E 94.66 是

Fig.6

Classification accuracy trend chart of CWRU"

Fig.7

Unoutstanding features classification accuracy trendchart of CWRU"

Fig.8

Classification accuracy trend chart of CUT-2"

Fig.9

Unoutstanding features classification accuracytrend chart of CUT-2"

Table 7

Comparison of failure identification rate before and after using of outstanding features"

%
数据集 分类方法 p q
CWRU SVM 89.58 97.37
CUT-2 SVM 87.76 95.38

Fig.10

Comparison of rate of features selected"

1 张美玲, 胡晓. 基于LCD和改进SVM的轴承故障诊断方法[J]. 测控技术与仪器仪表, 2016, 6, 81- 83.
ZHANG Meiling , HU Xiao . Fault diagnosis method of rolling bearing based on LCD and improved SVM[J]. Measurement Control Technology and Instruments, 2016, (6): 81- 83.
2 许凡, 方彦军, 张荣. 基于EEMD模糊熵的PCA-GG滚动轴承聚类故障诊断[J]. 计算机集成制造系统, 2016, 22 (11): 2631- 2642.
XU Fan , FANG Yanjun , ZHANG Rong . PCA-GG rolling bearing clustering fault diagnosis based on EEMD fuzzy entropy[J]. Computer Integrated Manufacturing System, 2016, 22 (11): 2631- 2642.
3 LI Shaobo , LIU Guokai , TANG Xianghong , et al. An nsemble deep convolutional neural network model with improved D-S evidence fusion for bearing fault diagnosis[J]. Sensors, 2017, 17 (8): 1729- 1748.
doi: 10.3390/s17081729
4 张钰, 陈珺, 王晓峰, 等. 随机森林在滚动轴承故障诊断中的应用[J]. 计算机工程与应用, 2017, (5): 1- 7.
ZHANG Yu , CHEN Jun , WANG Xiaofeng , et al. Application of randomforest on rolling element bearings fault diagnosis[J]. Computer Engineering and Applications, 2017, (5): 1- 7.
5 马萍, 张宏立, 范文慧. 基于局部与全局结构保持算法的滚动轴承故障诊断[J]. 机械工程学报, 2017, 53 (2): 20- 25.
MA Ping , ZHANG Hongli , FAN Wenhui . Fault diagnosis of rolling bearings based on local and global preserving embedding algorithm[J]. Joural of Mechanical Engineer, 2017, 53 (2): 20- 25.
6 VAKHARIA V , GUPTA VK , KANKAR PK . Efficient fault diagnosis of ball bearing using reliefF and random forest classifier[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2017, 39 (8): 2969- 2982.
doi: 10.1007/s40430-017-0717-9
7 朱瑜, 王殿, 王海洋. 基于EMD和信息熵的滚动轴承故障诊断[J]. 轴承, 2012, (6): 50- 53.
ZHU Yu , WANG Dian , WANG Haiyang . Rolling bearing fault diagnosis based on EMD and information entropy[J]. Bearings, 2012, (6): 50- 53.
8 许凡, 方彦军, 张荣. 基于EEMD模糊熵的PCA-GG滚动轴承聚类故障诊断[J]. 计算机集成制造系统, 2016, 22 (11): 2631- 2642.
XU Fan , FANG Yanjun , ZHANG Rong . PCA-GG rolling bearing clustering fault diagnosis based on EEMD fuzzy entropy[J]. Computer Integrated Manufacturing System, 2016, 22 (11): 2631- 2642.
9 杨宇, 潘海洋, 程军圣. 基于特征选择和RRVPMCD的滚动轴承故障诊断方法[J]. 振动工程学报, 2014, 27 (4): 629- 636.
doi: 10.3969/j.issn.1004-4523.2014.04.020
YANG Yu , PAN Haiyang , WEI Junsheng . The rolling bearing fault diagnosis method based on the feature selection and RRVPMCD[J]. Journal of Vibration Engineer, 2014, 27 (4): 629- 636.
doi: 10.3969/j.issn.1004-4523.2014.04.020
10 刘蕴哲, 胡金海, 任立通, 等. 基于特征选择与概率神经网络的轴承故障诊断研究[J]. 机械传动, 2016, 40 (10): 48- 53.
LIU Yunzhe , HU Jinhai , REN Litong , et al. Study on the bearing fault diagnosis based on feature selection and probabilistic neural network[J]. Journal of Mechanical Transmission, 2016, 40 (10): 48- 53.
11 苏祖强, 汤宝平, 姚金宝. 基于敏感特征选择与流形学习维数约简的故障诊断[J]. 振动与冲击, 2014, 33 (3): 70- 75.
doi: 10.3969/j.issn.1000-3835.2014.03.015
SU Zuqiang , TANG Baoping , YAO Jinbao . Fault diagnosis method based on sensitive feature selection and manifold learning dimension reduction[J]. Journal of Vibration and Shock, 2014, 33 (3): 70- 75.
doi: 10.3969/j.issn.1000-3835.2014.03.015
12 LI Yun , LI Tao , LIU Huan . Recent advances in feature selection and its applications[J]. Knowledge and Information System, 2017, 53 (3): 551- 577.
doi: 10.1007/s10115-017-1059-8
13 李敏, 卡米力·木依丁. 特征选择方法与算法的研究[J]. 计算机技术与发展, 2013, 23 (12): 16- 21.
LI Min , KAMIL Moydi . Research on feature selection methods and algorithms[J]. Computer Technology and Development, 2013, 23 (12): 16- 21.
14 王新, 闫文源. 基于变分模态分解和SVM的滚动轴承故障诊断[J]. 振动与冲击, 2017, 36 (18): 252- 256.
WANG Xin , YAN Wenyuan . Fault diagnosis of roller bearings based on the variational mode decomposition and SVM[J]. Journal of Vibration and Shock, 2017, 36 (18): 252- 256.
15 陈法法, 杨晶晶, 肖文荣, 等. Adaboost_SVM集成模型的滚动轴承早期故障诊断[J]. 机械科学与技术, 2018, 37 (2): 237- 243.
CHEN Fafa , YANG Jingjing , XIAO Wenrong , et al. Early fault diagnosis of rolling bearing based on ensemble model of adaboost SVM[J]. Mechanical Science and Technology for Aerospace Engineering, 2018, 37 (2): 237- 243.
16 黄勇, 郑春颖, 宋忠虎. 多类支持向量机算法综述[J]. 计算技术与自动化, 2005, (4): 61- 63.
doi: 10.3969/j.issn.1003-6199.2005.04.020
HUANG Yong , ZHENG Chunying , SONG Zhonghu . Multi-class support vector machines algorithm summarization[J]. Computing Technology and Automation, 2005, (4): 61- 63.
doi: 10.3969/j.issn.1003-6199.2005.04.020
17 肖晓, 张敏. 支持向量机多分类问题研究[J]. 淮海工学院学报(自然科学版), 2014, 23 (3): 28- 31.
XIAO Xiao , ZHANG Min . Research on the multi-class classification problem of support vector machine[J]. Journal of Huaihai Institute of Technology (Natural Science Edition), 2014, 23 (3): 28- 31.
18 付阳, 李昆仑. 支持向量机模型参数选择方法综述[J]. 电脑知识与技术, 2010, 6 (28): 8081- 80828085.
doi: 10.3969/j.issn.1009-3044.2010.28.077
FU Yang , LI Kunlun . A survey of model parameters selection method for support vector machines[J]. Computer Knowledge and Technology, 2010, 6 (28): 8081- 80828085.
doi: 10.3969/j.issn.1009-3044.2010.28.077
19 王玉静, 康守强, 张云, 等. 基于集合经验模态分解敏感固有模态函数选择算法的滚动轴承状态识别方法[J]. 电子与信息学报, 2014, 36 (3): 595- 600.
WANG Yujing , KANG Shouqiang , ZHANG Yun , et al. Condition recognition method of rolling bearing based on ensemble empirical mode decomposition sensitive intrinsic mode function selection algorithm[J]. Journal of Electronics and Information Technology, 2014, 36 (3): 595- 600.
[1] Caihui LIU,Qi ZHOU,Xiaowen YE. An intrusion detection model based on improved ReliefF algorithm [J]. Journal of Shandong University(Engineering Science), 2023, 53(2): 1-10.
[2] Haigen MIN,Yukun FANG,Xia WU,Wuqi WANG. Fault diagnosis of vehicle-to-vehicle communication in networked traffic environment [J]. Journal of Shandong University(Engineering Science), 2021, 51(6): 84-92.
[3] Yan PENG,Tingting FENG,Jie WANG. An integrated learning approach for O3 mass concentration prediction model [J]. Journal of Shandong University(Engineering Science), 2020, 50(4): 1-7.
[4] Xin MA,Xue WANG. Prediction of microRNA-binding residues based on Laplacian support vector machine and sequence information [J]. Journal of Shandong University(Engineering Science), 2020, 50(2): 76-82.
[5] Hong CHEN,Xiaofei YANG,Qing WAN,Yingcang MA. Multi-label feature selection algorithm based on correntropy andmanifold learning [J]. Journal of Shandong University(Engineering Science), 2018, 48(6): 27-36.
[6] Lianming MOU. Weighted k sub-convex-hull classifier based on adaptive feature selection [J]. Journal of Shandong University(Engineering Science), 2018, 48(5): 32-37.
[7] CHENG Xin, ZHANG Lin, HU Yefa, CHEN Qiang, LIANG Dian. Fault diagnosis of electromagnetic coil in active magnetic bearing based on current characteristics [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2018, 48(4): 94-101.
[8] CHENG Xin, LIU Han, WANG Bo, LIANG Dian, CHEN Qiang. A fault-tolerant control architecture for active magnetic bearing based on dual core processor [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2018, 48(2): 72-80.
[9] WANG Xiuqing, ZENG Hui, XIE Fei, LYU Feng. Fault diagnosis for manipulators based on Spiking neural networks [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 15-21.
[10] SONG Yang, ZHONG Maiying. Fault isolability analysis based on improved distance similarity [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 103-109.
[11] LI Wei, WANG Kehong, CAO Huichao. A fault filtering method based on an improved extended state filter for nonlinear system [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 7-14.
[12] MAO Haijie, LI Wei, WANG Kehong, FENG Xiaolin. Sensor fault tolerant switch strategy for multi-motor synchronous system based on ADRC [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 64-70.
[13] QIU Lu, YE Yinzhong, JIANG Chundi. Fault diagnostic method for micro-grid based on wavelet singularity entropy and SOM neural network [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 118-122.
[14] XIE Xiaolong, JIANG Bin, LIU Jianwei, JIANG Yinhang. Fault diagnosis and fault tolerant control based on sliding mode observer for speed sensor in asynchronous motor [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 210-214.
[15] WANG Mengyuan, ZHANG Xiong, MA Liang, PENG Kaixiang. Fault diagnosis for industrial processes based on causal topological graph [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2017, 47(5): 187-194.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] LI Kan . Empolder and implement of the embedded weld control system[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(4): 37 -41 .
[2] SUN Dianzhu, ZHU Changzhi, LI Yanrui. [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2009, 39(1): 84 -86 .
[3] . [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2009, 39(1): 92 -95 .
[4] SHI Lai-shun,WAN Zhong-yi . Synthesis and performance evaluation of a novel betaine-type asphalt emulsifier[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(4): 112 -115 .
[5] YU Jia yuan1, TIAN Jin ting1, ZHU Qiang zhong2. Computational intelligence and its application in psychology[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2009, 39(1): 1 -5 .
[6] HU Tian-liang,LI Peng,ZHANG Cheng-rui,ZUO Yi . Design of a QEP decode counter based on VHDL[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(3): 10 -13 .
[7] QU Yan-peng,CHEN Song-ying,LI Chun-feng,WANG Xiao-peng,TENG Shu-ge . [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2006, 36(4): 16 -20 .
[8] ZOU Feifei,GUAN Xiaojun,HAN Zhenqiang,SHEN Xiaomin,MA Xiaofei ,LIU Yunteng . hermal simulating experiment and FEM simulation of dynamic recrystallization of 09CuPTiRE steel[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2006, 36(5): 17 -20 .
[9] ZHANG Ai-juan. Synthesis of bone-like hydroxyapatite in simulated body fluid[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2010, 40(3): 86 -90 .
[10] WANG Ru-gui,CAI Gan-wei . Sub-harmonic resonance analysis of 2-DOF controllable plane linkage mechanism electromechanical coupling system[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2008, 38(3): 58 -63 .
Copyright 2018 © Editorial office of Journal of Shandong University (Engineering Science)
Supported by Beijing Magtech Co.Ltd