Fault Diagnosis Approach of Gear based on Two Features and Least Squares Support Vector Machine

Qin Bo; Yang Yunzhong; Chen Min; Guo Wei; Liu Yongliang; Wang Jianguo

doi:10.16578/j.issn.1004.2539.2016.06.028

您当前的位置：

首页 >

文章列表页 >

Fault Diagnosis Approach of Gear based on Two Features and Least Squares Support Vector Machine

更新时间：2022-10-20

- Fault Diagnosis Approach of Gear based on Two Features and Least Squares Support Vector Machine
- Vol. 40, Issue 6, Pages: 126-131(2016)
- 作者机构：
  
  1. 内蒙古科技大学机械工程学院
  2. 浙江大学信息学部控制科学学院
  3. 包头钢铁(集团)有限公司设备动力部
- 作者简介：
- 基金信息：
- DOI：10.16578/j.issn.1004.2539.2016.06.028
  CLC：
扫描看全文
[1]秦波,杨云中,陈敏,郭伟,刘永亮,王建国.基于两类特征和最小二乘支持向量机的齿轮故障诊断方法[J].机械传动,2016,40(06):126-131.

Qin Bo, Yang Yunzhong, Chen Min, et al. Fault Diagnosis Approach of Gear based on Two Features and Least Squares Support Vector Machine. [J]. 40(6):126-131(2016)
[1]秦波,杨云中,陈敏,郭伟,刘永亮,王建国.基于两类特征和最小二乘支持向量机的齿轮故障诊断方法[J].机械传动,2016,40(06):126-131. DOI： 10.16578/j.issn.1004.2539.2016.06.028.

Qin Bo, Yang Yunzhong, Chen Min, et al. Fault Diagnosis Approach of Gear based on Two Features and Least Squares Support Vector Machine. [J]. 40(6):126-131(2016) DOI： 10.16578/j.issn.1004.2539.2016.06.028.

摘要

针对齿轮振动信号非线性非平稳特性,为避免传统"时-频"分析方法在表征设备状态时的不足和样本数量少易造成故障辨识模型"欠学习"的问题,提出一种基于峭度、本征模式分量（Intrinsic mode function,IMF）能量两类特征和最小二乘支持向量机（Least squares support vector machine,LS-SVM）的齿轮故障诊断方法。首先,对所测齿轮振动信号在集合经验模式分解（Ensemble empirical mode decomposition,EEMD）的基础上提取有效IMF分量计算其能量特征和峭度值,据此构建时频域两类特征向量;其次,将融合后的齿轮正常、齿根裂纹、断齿3种状态下的时频域两类特征向量作为输入,基于LS-SVM建立齿轮故障诊断模型,进行齿轮故障识别。实验结果表明,该方法能够准确地识别齿轮的工作状态,与基于BP、SVM的故障诊断模型相比,其具有更高的识别率,为齿轮状态识别和故障诊断提供了一种新途径。

Abstract

Aiming at the Gear vibration signals have the nonlinear and non-stationary characteristics,to avoid the disadvantages of traditional time and frequency domain method in the characterization of the state of equipment and failure identification model "less learning"problem caused by small sample size,the gearbox fault diagnosis method based on kurtosis and intrinsic mode function（ IMF） energy feature and least squares support vector machine（ LS-SVM） is proposed. Firstly,by using ensemble empirical mode decomposition（ EEMD）,the collected gear vibration signal is decomposed,on this basis,the IMF components which contain major fault information are extracted and its energy feature and kurtosis are calculated,and the time-frequency domain two kinds of the feature vector are constructed. Secondly,taking the fusion feature vectors of three conditions of normal,the root crack and broken as input,the gearbox fault type identification is conducted based on the LS-SVM. The experiment results show that the gear working state can be accurately identified by this method. It has higher efficiency of fault identification compared with the BP neural network and SVM model and a new way for the gear fault diagnosis is provided.

关键词

本征模式分量能量峭度最小二乘支持向量机齿轮故障诊断

Keywords

IMF energyKurtosisLeast squares support vector machineGearFault diagnosis

references

Views

181

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

SVM Diagnosis of Gear Fault based on Improved Feature Selection Method

Application of Deep Support Vector Machine in Gear Fault Diagnosis

Research on Gear Fault Diagnosis Method based on Local Characteristic Scale Decomposition and Mathematical Morphological Filtering

Gear Fault Diagnosis based on EEMD and Choi-Williams Distribution

Gear Fault Diagnosis Method based on Variational Modal Decomposition and Symbol Entropy

Related Author

No data

Related Institution

School of Information Engineering， ZhengZhou Tourism College

School of Mechanical Engineering， Jiangnan University

School of Mechanical Engineering， Donghua University

Coastal Defense College， Naval Aviation University

School of Mechanical and Power Engineering， Henan Polytechnic University

⁰