Improved Bearing Fault Diagnosis by Feature Extraction Based on GLCM, Fusion of Selection Methods, and Multiclass-Naïve Bayes Classification  

作　　者：Mireille Pouyap Laurent Bitjoka Etienne Mfoumou Denis Toko Mireille Pouyap;Laurent Bitjoka;Etienne Mfoumou;Denis Toko(Energy, Signal, Imaging and Automation Laboratory (LESIA), University of Ngaoundere, Yaounde, Cameroon;Division of Applied Research & Innovation, Nova Scotia Community College, Dartmouth, Canada;Department of Mechanical Engineering, College of Technology, University of Buea, Buea, Cameroon)

机构地区：[1]Energy, Signal, Imaging and Automation Laboratory (LESIA), University of Ngaoundere, Yaounde, Cameroon [2]Division of Applied Research & Innovation, Nova Scotia Community College, Dartmouth, Canada [3]Department of Mechanical Engineering, College of Technology, University of Buea, Buea, Cameroon

出　　处：《Journal of Signal and Information Processing》2021年第4期71-85,共15页信号与信息处理（英文）

摘　　要：<span style="font-family:Verdana;">The presence of bearing faults reduces the efficiency of rotating machines and thus increases energy consumption or even the total stoppage of the machine. </span><span style="font-family:Verdana;">It becomes essential to correctly diagnose the fault caused by the bearing.</span><span style="font-family:Verdana;"> Hence the importance of determining an effective features extraction method that best describes the fault. The vision of this paper is to merge the features selection methods in order to define the most relevant featuresin the texture </span><span style="font-family:Verdana;">of the vibration signal images. In this study, the Gray Level Co-occurrence </span><span style="font-family:Verdana;">Matrix (GLCM) in texture analysis is applied on the vibration signal represented in images. Features</span><span><span><span style="font-family:;" "=""> </span></span></span><span><span><span style="font-family:;" "=""><span style="font-family:Verdana;">selection based on the merge of PCA (Principal component Analysis) method and SFE (Sequential Features Extraction) method is </span><span style="font-family:Verdana;">done to obtain the most relevant features. The multiclass-Na<span style="white-space:nowrap;">?</span>ve Bayesclassifi</span><span style="font-family:Verdana;">er is used to test the proposed approach. The success rate of this classification is 98.27%. The relevant features obtained give promising results and are more efficient than the methods observed in the literature.</span></span></span></span><span style="font-family:Verdana;">The presence of bearing faults reduces the efficiency of rotating machines and thus increases energy consumption or even the total stoppage of the machine. </span><span style="font-family:Verdana;">It becomes essential to correctly diagnose the fault caused by the bearing.</span><span style="font-family:Verdana;"> Hence the importance of determining an effective features extraction method that best describes the fault. The vision of this paper is to merge the features selection methods in order to define the most relevant featuresin the texture </span><span style="font-family:Verdana;">of the vibration signal images. In this study, the Gray Level Co-occurrence </span><span style="font-family:Verdana;">Matrix (GLCM) in texture analysis is applied on the vibration signal represented in images. Features</span><span><span><span style="font-family:;" "=""> </span></span></span><span><span><span style="font-family:;" "=""><span style="font-family:Verdana;">selection based on the merge of PCA (Principal component Analysis) method and SFE (Sequential Features Extraction) method is </span><span style="font-family:Verdana;">done to obtain the most relevant features. The multiclass-Na<span style="white-space:nowrap;">?</span>ve Bayesclassifi</span><span style="font-family:Verdana;">er is used to test the proposed approach. The success rate of this classification is 98.27%. The relevant features obtained give promising results and are more efficient than the methods observed in the literature.</span></span></span></span>

关 键 词：GLCM PCA SFE Naïve Bayes Relevant Features 

分 类 号：O17[理学—数学]