《机电工程》杂志社科技期刊核心期刊论文投稿机电一体化机械、仪表电工自动化、计算机浙江大学主办

Introduction

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

Gear fault identification method based on CSAEMD-KECA and angular structure distance

Published:2023-03-23 author:GAO Qing-yun, GUO Li, CHEN Chang-hua. Browse: 1400 Check PDF documents

Gear fault identification method based on CSAEMDKECA

and angular structure distance

GAO Qing-yun1, GUO Li2, CHEN Chang-hua3

(1.Hangzhou Vocational and Technical College, Hangzhou 310018, China; 2.School of Mechanical and

Electrical and Vehicle Engineering, Chongqing Jiaotong University, Chongqing 400074, China;

3.Chongqing Changjiang Bearing Co.,Ltd., Chongqing 401336, China)

Abstract: As an important part of the mechanical transmission system, gears often operate under variable speed and load conditions. The directly collected gear fault signals (original signals) often have strong background noise. Aiming at the problem that the noise signal in the original signal interfered with the gear fault pattern recognition and the traditional fault identification method had a low accuracy rate, a gear fault identification method based on the complementary sine assisted empirical mode decompositionkernel entropy component analysis (CSAEMD-KECA) and angular structure distance was proposed. Firstly, the gear fault signal was decomposed and reconstructed by the complementary sine assisted empirical mode decomposition (CSAEMD) method to remove the noise components in the signal. Then, the feature extraction of the signal decomposed and reconstructed by CSAEMD was carried out by using the kernel entropy component analysis (KECA). The three eigenvectors with larger contribution to the Rayleigh entropy of the sample (the signal after CSAEMD decomposition and reconstruction) was selected as the projection vector, and the sample data was projected to the projection vector to form a feature dataset. Finally,a fault simulation test bench was built to verify the feasibility of the above method.The clustering method of angular structure distance was used to perform cluster analysis on the feature dataset. The research results show that the effective experiments using the experimental bench data can accurately identify various faults of gears, and the clustering accuracy rate reaches 98.3%. The results verify the effectiveness of the proposed method in gear fault identification.

Key words: mechanical transmission system; gearfault diagnosis; complementary sine assisted empirical mode decomposition(CSAEMD); kernel entropy component analysis(KECA); cluster analysis; signal decomposition and reconstruction; signal feature extraction

Advanced Search

Online Office System

Chinese Core Periodicals
Chinese Sci-tech Core Periodicals
SA, INSPEC Indexed
CSA: T Indexed
UPD:Indexed