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 view of the non-linear and non-stationary characteristics of gearbox fault signals and the insufficient feature extraction of current methods, the integration of different coarse-graining methods and the application of multi-channel signal processing methods in fuzzy entropy algorithm were studied. A new feature extraction method called ensemble refined composite multivariate multi-scale fuzzy entropy (ERCmvMFE) was proposed. Based on ERCmvMFE, combining with t-SNE and kernel extreme learning machine optimized by artificial fish swarm optimization algorithm (AFSA-KELM), a new gear fault comprehensive diagnosis method was proposed. Firstly, the fuzzy entropy algorithm(FEA)was improved by using the integration method of multiple coarse-grained methods and the multi-channel signal processing method, and the initial fault features were extracted. Then, t-SNE was used to compress the original fault features to obtain low-dimensional sensitive features,the low-dimensional fault features were input to the AFSA-KELM classifier for fault classification and identification. Finally,in order to test the feature extraction performance of the ERCmvMFE method, relevant experiments were carried out using the QPZZ-II rotating machinery fault simulation test platform.The experiment results show that the proposed method can reliably diagnose different types of gearbox faults. The average recognition accuracy of 20 recognition experiments for five working conditions of gearbox can reach 98.92%, and the standard deviation is 0.956. The recognition accuracy and stability are better than the comparison method. The results show that the ERCmvMFE algorithm can more fully extract the fault characteristics of the gearbox. Therefore,the fault diagnosis method based on this feature extraction method has higher fault recognition accuracy.
Key words: ensemble refined composite multivariate multiscale fuzzy entropy (ERCmvMFE); kernel extreme learning machine optimized by artificial fish swarm optimization algorithm (AFSA-KELM); tdistributed stochastic neighbor embedding(t-SNE); feature extraction; coarse grained process; multichannel signal processing; fault classification and identification
-
- Chinese Core Periodicals
-
- Chinese Sci-tech Core Periodicals
-
- SA, INSPEC Indexed
-
- CSA: T Indexed
-
- UPD:Indexed
-
2010 Zhejiang Information Institute of Mechinery Industry
Technical Support：Hangzhou Bory science and technology
