FAULT DIAGNOSIS METHOD OF ROLLING BEARING BASED ON ICEEMD-FastICA

MA WeiPing; HONG KunYue; AN Ning; SONG YuZhou

doi:10.16579/j.issn.1001.9669.2024.02.004

您当前的位置：

首页 >

文章列表页 >

FAULT DIAGNOSIS METHOD OF ROLLING BEARING BASED ON ICEEMD-FastICA

更新时间：2024-07-05

- FAULT DIAGNOSIS METHOD OF ROLLING BEARING BASED ON ICEEMD-FastICA
- Journal of Mechanical Strength Vol. 46, Issue 2, Pages: 281-285(2024)
- 作者机构：
  
  1. 郑州机械研究所有限公司
  2. 机械工业第六设计研究院有限公司
- 作者简介：
- 基金信息：
  
  The project supported by the China Academy of Mechanical Science Foundation Project (No. AZDJJ5032103).
- DOI：10.16579/j.issn.1001.9669.2024.02.004
  CLC： TH165+.3
- Published：15 April 2024，
- 稿件说明：
移动端阅览
MA WeiPing, HONG KunYue, AN Ning, et al. FAULT DIAGNOSIS METHOD OF ROLLING BEARING BASED ON ICEEMD-FastICA. [J]. Journal of Mechanical Strength , 2024,46(2):281-285.
DOI：

MA WeiPing, HONG KunYue, AN Ning, et al. FAULT DIAGNOSIS METHOD OF ROLLING BEARING BASED ON ICEEMD-FastICA. [J]. Journal of Mechanical Strength , 2024,46(2):281-285. DOI： 10.16579/j.issn.1001.9669.2024.02.004.

摘要

针对滚动轴承早期故障特征信号提取困难的问题，提出了一种改进完备集成经验模态分解（Improved Complete Ensemble Empirical Mode Decomposition，ICEEMD）和独立分量分析（Independent Component Analysis，ICA）联合故障诊断方法。该方法利用峭度准则将经ICEEMD得到的固有模态分量（Intrinsic Mode Function，IMF）重构后结合快速独立分量分析（Fast Independent Component Analysis，FastICA）进行降噪解混，明显降低被测信号中的噪声，并且在故障特征频率处能量幅值取得最大值，便于辨识故障特征。通过试验研究分析，表明该方法可以明显降低噪声干扰，突出故障频率成分。和ICEEMD与包络谱结合的方法对比，信噪比提高了29.54%，能更准确地识别故障特征，达到对滚动轴承故障的判别需求，从而为轴承故障特征提取提供了一种新思路。

Abstract

In response to the difficulty in extracting early fault feature signals of rolling bearings

a joint fault diagnosis method based on Improved Complete Ensemble Empirical Mode Decomposition (ICEEMD) and Independent Component Analysis(ICA) was proposed. This method utilized the kurtosis criterion to reconstruct the Intrinsic Mode Function (IMF) obtained from ICEEMD and combined it with Fast Independent Component Analysis (FastICA) for noise reduction and unmixing

significantly reducing the noise in the measured signals. The maximum energy amplitude was obtained at the fault feature frequency

making it easy to identify fault features. Through experimental research and analysis

it is shown that this method can significantly reduce noise interference and highlight fault frequency components. Compared with the method combining ICEEMD and envelope spectrum

the signal-to-noise ratio is inereased by 29.54%

which can more accurately identify fault features and meet the discrimination requirements for rolling bearing faults

thus providing a new approach for bearing fault feature extraction.

关键词

改进完备集成经验模态分解盲源分离独立分量分析故障诊断降噪

Keywords

Improved complete ensemble empirical mode decompositionBlind source separationIndependent component analysisFault diagnosisNoise reduction

references

Views

276

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

RESEARCH ON ROLLING BEARING FAULT DIAGNOSIS BASED ON IMPROVED HHO-LSTM

ROLLING BEARING FAULT DIAGNOSIS BASED ON ADAPTIVE FEATURE SELECTION k-SUB CONVEX HULL

基于SAVMD与CNN结合的轴承故障诊断方法

改进MOMEDA方法及其在滚动轴承故障特征增强中的应用

基于XGBoost的冷水机组不平衡数据故障诊断

Related Author

ZHU SiJia

SHAO LiangShan

CHENG JunSheng

YANG Yu

WU ZhanTao

HU AiRu

贾博

杜刚

Related Institution

System Engineering Institute,Liaoning Technical University

School of Business Administration,Liaoning Technical University

College of Mechanical and Vehicle Engineering, Hunan University

Hunan Provincial Key Laboratory of Equipment Service Quality Assurance, Hunan University

武汉理工大学机电工程学院

⁰