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

RollingbearingfaultdiagnosismethodbasedonCCGANandResNet34
Published:2023-08-14 author:LUOYao-pu.WANGYan-xue.LIMeng. Browse: 335 Check PDF documents
Rolling bearing fault diagnosis method based on CCGAN and ResNet34 
LUO Yao ̄pu 1. WANG Yan ̄xue1.2. LI Meng1 
(1. School of Mechanical ̄Electronic and Vehicle Engineering. Beijing University of Civil Engineering and Architecture. Beijing 102616, China;2. Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles. Beijing University of Civil Engineering and Architecture. Beijing 100044.,China) 
Abstract: Aiming at the problem that it is difficult for many algorithms to accurately identify different faults due to the small sample unbalanced rolling bearing data collected in the industrial process. a deep neural network fault diagnosis method based on the conditional convolution generates adversarial networks (CCGAN) and ResNet34 was proposed. Firstly. the vibration signal data of the rolling bearing were collected. and the vibration signal was converted into gray image and the data characteristics were enhanced. Then. the CCGAN network was used to learn the features of the original small sample data and expand the small sample unbalanced data set. Finally. on the 
basis of data expansion and feature enhancement of vibration signals of rolling bearings. ResNet34 deep network was used to diagnose and classify the small sample unbalance fault of one ̄dimensional vibration signals. The experimental results show that as the small sample unbalanced data set gradually expands to the multi ̄dimensional balanced data set. the accuracy of the proposed method in different data sets is effectively improved. and the classification accuracy reaches 99. 5%. which proves that the feature extraction capability of the proposed 
method is superior to the typical machine learning and deep learning neural network. thus verifying the advantages of the method in the small sample unbalanced fault diagnosis. 
Key words: small sample fault diagnosis;data expansion. deep learning; generating adversarial networks(GAN); residual structure(RS);conditional convolution generates adversarial networks (CCGAN);improved feature extraction enhancements (IFEE) method
  • Chinese Core Periodicals
  • Chinese Sci-tech Core Periodicals
  • SA, INSPEC Indexed
  • CSA: T Indexed
  • UPD:Indexed

Copyright 2010 Zhejiang Information Institute of Mechinery Industry All Rights Reserved

Technical Support:Hangzhou Bory science and technology

You are 1895221 visit this site