Founded in 1971 >
Chinese Sci-tech Core Periodicals >
British Science Abstracts (SA, INSPEC) Indexed Journals >
United States, Cambridge Scientific Abstract: Technology (CSA: T) Indexed Journals >
United States, Ulrich's Periodicals Directory(UPD)Indexed Journals >
United States, Cambridge Scientific Abstract: Natural Science (CSA: NS) Indexed Journals >
Poland ,Index of Copernicus(IC) Indexed Journals >
International Standard Serial Number:
ISSN 1001-4551
Sponsor:
Zhejiang University;
Zhejiang Machinery and Electrical Group
Edited by:
Editorial of Journal of Mechanical & Electrical Engineering
Chief Editor:
ZHAO Qun
Vice Chief Editor:
TANG ren-zhong,
LUO Xiang-yang
Tel:
86-571-87041360,87239525
Fax:
86-571-87239571
Add:
No.9 Gaoguannong,Daxue Road,Hangzhou,China
P.C:
310009
E-mail:
meem_contribute@163.com
Abstract: In order to solve the problem that support matrix machine (SMM) lacked necessary probability information in classification modeling, resulting in its unclear sparsity and robustness, a new matrix classifier called relevance support matrix machine (RSMM) was proposed according to Bayesian theory framework. In RSMM, firstly the multi-classification objective function was established with the matrix as the modeling element, the structured information between rows and columns of the input matrix was made full use of to establish accurate prediction model. Then, Bayesian learning framework was used to impose a conditional probability distribution constraint on model parameters to obtain sparse solution space. The kernel function of the RSMM method was not restricted by Mercer condition, and the probability and statistical information between different categories could be obtained.The prior probability was introduced into the model weight setting to make the RSMM model have the feature of sparsity, and the uncertain samples were classified. Finally, the rolling bearing fault classification experiment was carried out, and the rolling bearing data set was used to test the performance of this method.The results show that RSMM can accurately classify uncertain samples by using Bayesian learning framework and a prior probability. At the same time, RSMM can make full use of the structured information of samples.Comparing with SMM and its improved algorithms, RSMM improves the overall recognition rate by 2%~8%, which proves that RSMM has superior classification performance in roller bearing fault diagnosis.
Key words: rolling bearing; fault diagnosis; relevance support matrix machine(RSMM); Bayesian framework
CHEN Ying, CHEN Mu-rong. Fault diagnosis method of rolling bearing based on relevance support matrix machine[J].Journal of Mechanical & Electrical Engineering, 2021,38(12):1592-1598.