一种基于改进VMD和UMAP的滚动轴承故障特征提取方法

刘俊利; 缪炳荣; 张盈; 李永健; 黄仲

doi:10.16578/j.issn.1004.2539.2023.06.019

您当前的位置：

首页 >

文章列表页 >

一种基于改进VMD和UMAP的滚动轴承故障特征提取方法

开发应用 | 更新时间：2023-08-14

- 一种基于改进VMD和UMAP的滚动轴承故障特征提取方法
- A Fault Feature Extraction Method of Rolling Bearings Based on Optimized VMD and UMAP
- 机械传动 2023年47卷第6期页码：130-138
- 作者机构：
  
  1.西南交通大学牵引动力国家重点实验室，四川成都 610031
  2.五邑大学轨道交通学院，广东江门 529020
- 作者简介：
  
  刘俊利（2000— ），女，重庆人，硕士研究生；研究方向为机器学习和结构故障诊断问题研究；ljl1610542903@163.com。
- 基金信息：
  
  国家自然科学基金项目(51775456┫牵引动力国家重点实验室自主研究课题┣2019TPL_T03)
- DOI：10.16578/j.issn.1004.2539.2023.06.019
  中图分类号：
- 纸质出版日期：2023-06-15，
  
  收稿日期：2022-04-26，
  
  修回日期：2022-05-19，
扫描看全文
刘俊利,缪炳荣,张盈等.一种基于改进VMD和UMAP的滚动轴承故障特征提取方法[J].机械传动,2023,47(06):130-138.

Liu Junli,Miao Bingrong,Zhang Ying,et al.A Fault Feature Extraction Method of Rolling Bearings Based on Optimized VMD and UMAP[J].Journal of Mechanical Transmission,2023,47(06):130-138.
刘俊利,缪炳荣,张盈等.一种基于改进VMD和UMAP的滚动轴承故障特征提取方法[J].机械传动,2023,47(06):130-138. DOI： 10.16578/j.issn.1004.2539.2023.06.019.

Liu Junli,Miao Bingrong,Zhang Ying,et al.A Fault Feature Extraction Method of Rolling Bearings Based on Optimized VMD and UMAP[J].Journal of Mechanical Transmission,2023,47(06):130-138. DOI： 10.16578/j.issn.1004.2539.2023.06.019.

摘要

针对滚动轴承振动信号难以提取低维度敏感特征的问题，提出了一种基于逻辑回归优化变分模态分解（Variational Mode Decomposition，VMD）及均匀流形逼近和投影（Uniform Manifold Approximation and Projection，UMAP）的滚动轴承故障特征提取方法。首先，通过移动窗方法进行原始数据的样本划分，完成训练集以及测试集的构建；其次，随机抽取训练集部分数据进行不同分解模态数下的VMD，并对各层子信号进行特征提取，完成多个特征集的构建；然后，通过逻辑回归分别计算各特征集中各特征与标签的复相关系数，以确定变分模态分解的分解模态数和高度相关特征，并将此应用于训练集和测试集，得到高维特征数据集；最后，采用UMAP降维，获取具有高判别性的低维度特征，完成最终特征集构建。以3种常用智能算法的识别准确率及测试特征集中类内余弦距离和类间余弦距离的比值作为评价指标，结果表明，该方法不仅能实现多种轴承故障特征的有效提取，而且抗噪性良好，对于实际轴承故障诊断时的特征提取具有一定的参考价值。

Abstract

In view of the problem of difficulties in extracting low-dimensional sensitive features from rolling bearing vibration signals

a rolling bearing fault feature extraction method based on logistic regression optimization variation mode decomposition(VMD) and uniform manifold approximation and projection(UMAP) is proposed. Firstly

the method divides the original data into samples by the moving window method

and completes the construction of the training set as well as the test set. Secondly

a part of the training set is randomly selected for VMD with different number of modal decompositions

and features are extracted for each layer of sub-signals to complete the construction of multiple feature sets. Then

the complex correlation coefficients between each feature and the label in each feature set are calculated by logistic regression to determine the number of modal decompositions and highly correlated features

which are applied to the training set and the test set to obtain the high-dimensional feature data set. Finally

UMAP is used to obtain low-dimensional features with high discriminative power to complete the final feature set construction. Using the recognition accuracy of three commonly used intelligent algorithm and the ratio of intra-class cosine distance and inter-class cosine distance in the tested feature set as evaluation indexes

the results show that the method not only achieves effectively extract features of various bearing failures

but also has good noise immunity

which is of certain reference value for the feature extraction in the practical bearing fault diagnosis.

关键词

故障诊断滚动轴承变分模态分解逻辑回归均匀流形逼近和投影

Keywords

Fault diagnosisRolling bearingVMDLogistic regressionUMAP

references

HE C，NIU P，YANG R，et al.Incipient rolling element bearing weak fault feature extraction based on adaptive second-order stochastic resonance incorporated by mode decomposition［J］.Measurement，2019，145：687-701.

WANG Z，CHEN H，YAO L，et al.An effective multi-channel fault diagnosis approach for rotating machinery based on multivariate generalized refined composite multi-scale sample entropy［J］.Nonlinear Dynamics，2021，106（3）：1-24.

ZHANG X，MIAO Q，ZHANG H，et al.A parameter-adaptive VMD method based on grasshopper optimization algorithm to analyze vibration signals from rotating machinery［J］.Mechanical Systems and Signal Processing，2018，108：58-72.

TIAN S，BIAN X，TANG Z，et al.Fault diagnosis of gas pressure regulators based on CEEMDAN and feature clustering［J］.IEEE Access，2019，7：132492-132502.

LI H，LIU T，WU X，et al.Application of EEMD and improved frequency band entropy in bearing fault feature extraction［J］.ISA Transactions，2019，88：170-185.

MIAO Y，ZHAO M，LIN J.Identification of mechanical compound-fault based on the improved parameter-adaptive variational mode decomposition［J］.ISA Transactions，2018，84：82-95.

ZHENG J，YUAN Y，ZOU L，et al.Study on a novel fault diagnosis method based on VMD and BLM［J］.Symmetry，2019，11（6）：747.

CHENG H，ZHANG Y，LU W，et al.A bearing fault diagnosis method based on VMD-SVD and fuzzy clustering［J］.International Journal of Pattern Recognition and Artificial Intelligence，2019，33（12）：25.

YE M，YAN X，JIA M.Rolling bearing fault diagnosis based on VMD-MPE and PSO-SVM［J］.Entropy，2021，23（6）：762.

BIE F，HOROSHENKOV K V，QIAN J，et al.An approach for the impact feature extraction method based on improved modal decomposition and singular value analysis［J］.Journal of Vibration and Control，2019，25（5）：1096-1108.

LI J，CHEN W，HAN K，et al.Fault diagnosis of rolling bearing based on GA-VMD and improved WOA-LSSVM［J］.IEEE Access，2020，8：166753-166767.

DING J，XIAO D，LI X.Gear fault diagnosis based on genetic mutation particle swarm optimization VMD and probabilistic neural network algorithm［J］.IEEE Access，2020，8：18456-18474.

马萍，张宏立，范文慧.基于局部与全局结构保持算法的滚动轴承故障诊断［J］.机械工程学报，2017，53（2）：20-25.

MA Ping，ZHANG Hongli，FAN Wenhui.Fault diagnosis of rolling bearings based on local and global preserving embedding algorithm［J］.Journal of Mechanical Engineering，2017，53（2）：20-25.

ZHENG J，JIANG Z，PAN H.Sigmoid-based refined composite multiscale fuzzy entropy and t-SNE based fault diagnosis approach for rolling bearing［J］.Measurement，2018，129：332-342.

WANG X，FAN W，LI X，et al.Weak degradation characteristics analysis of UAV motors based on Laplacian Eigenmaps and variational mode decomposition［J］.Sensors，2019，19（3）：524.

AKPUDO U E，HUR J W.Intelligent solenoid pump fault detection based on MFCC features，LLE and SVM［C］//2020 International Conference on Artificial Intelligence in Information and Communication （ICAIIC）.IEEE，2020：404-408.

GAO S，ZHANG S，ZHANG Y，et al.Operational reliability evaluation and prediction of rolling bearing based on isometric mapping and NoCuSa-LSSVM［J］.Reliability Engineering & System Safety，2020，201：106968.

MCLNNES L，HEALY J，SAUL N，et al.UMAP：uniform manifold approximation and projection［J］.Journal of Open Source Software，2018，3（29）：861.

DRAGOMIRETSKIY K，ZOSSO D.Variational mode decomposition［J］.IEEE Transactions on Signal Processing，2014，62（3）：531-544.

LIANG P，SONG X，WANG S，et al.Remaining useful life prediction for rolling bearings using correlation coefficient and Kullback-Leibler divergence feature selection［J］.Measurement Science and Technology，2021，33（2）：025005.

KUMARI K，DEY P，KUMAR C，et al.UMAP and LSTM based fire status and explosibility prediction for sealed-off area in underground coal mine［J］.Process Safety and Environmental Protection，2021，146： 837-852.

LIU R，YANG B，ZIO E，et al.Artificial intelligence for fault diagnosis of rotating machinery：a review［J］.Mechanical Systems and Signal Processing，2018，108：33-47.

PENG B S，XIA H，LIU Y K，et al.Research on intelligent fault diagnosis method for nuclear power plant based on correlation analysis and deep belief network［J］.Progress in Nuclear Energy，2018，108：419-427.

ANOWAR F，SADAOUI S，SELIM B.Conceptual and empirical comparison of dimensionality reduction algorithms （PCA，KPCA，LDA，MDS，SVD，LLE，ISOMAP，LE，ICA，t-SNE）［J］.Computer Science Review，2021，40：100378.

HAN T，ZHANG L，YIN Z，et al.Rolling bearing fault diagnosis with combined convolutional neural networks and support vector machine［J］.Measurement，2021，177：109022.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于时间子序列的轴承特征提取方法

基于变分模态分解和卷积神经网络融合的滚动轴承故障诊断方法

SVD-LMD联合降噪和TEO的滚动轴承故障诊断

基于敏感分量与MCPG的轴承故障诊断方法

基于同步挤压提取变换的滚动轴承故障诊断研究