[1]梁雄鹤,陈 珊,魏 豪?,等.基于最优 IMF 分量和 K-SVD 的滚动轴承故障声音信号特征提取[J].机械与电子,2022,(02):8-12.
　LIANG Xionghe,CHEN Shan,WEI Hao,et al.Extraction of Rolling Bearing Fault Sound Signal Features Based on Optimal IMF Components and K-SVD[J].Machinery & Electronics,2022,(02):8-12.
点击复制

基于最优 IMF 分量和 K-SVD 的滚动轴承故障声音信号特征提取()
分享到：

机械与电子[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2022年02期
页码:
8-12
栏目:
设计与研究
出版日期:
2022-02-22

文章信息/Info

Title:
Extraction of Rolling Bearing Fault Sound Signal Features Based on Optimal IMF Components and K-SVD
文章编号:
1001-2257 ( 2022 ) 02-0008-05
作者:
梁雄鹤; 陈 珊; 魏 豪?; 张丽洁; 权 伟
西安工程大学机电工程学院,陕西 西安 710048
Author(s):
LIANG Xionghe ;  CHEN Shan ;  WEI Hao ;  ZHANG Lijie ;  QUAN Wei
( School of Mechanical and Electrical Engineering , Xi ’ an Polytechnic University , Xi ’ an 710048 , China )
关键词:
声音信号; SAF 指标; 最优 IMF 分量;  K-SVD ; 信噪比
Keywords:
sound signal ;  SAF index ;  optimal IMF components ;  K-SVD ;  SNR
分类号:
TH133.3
文献标志码:
A
摘要:
针对滚动轴承声音信号中周期性冲击故障特征难提取的问题,提出了基于最优 IMF 分量与 K-SVD 字典学习相结合的轴承故障特征提取方法。首先,利用 VMD 分解原始信号获得一系列 IMF 分量;其次,利用 SAF 指标自适应选取最优 IMF 分量,并作为训练信号;最后,利用 K-SVD 字典学习方法训练出字典库,通过正交匹配追踪算法( OMP )对原始信号处理得到稀疏信号,并对稀疏信号进行包络谱分析。仿真及实验结果表明,对比传统 K-SVD 字典学习方法,该方法得到的稀疏信号信噪比( SNR )更高,能更准确地提取滚动轴承周期性冲击,增强了轴承故障特征。
Abstract:
For the problem of difficult extraction of periodic pulse fault features in rolling bearing sound signals , a bearing fault feature extraction method based on the combination of optimal modal components and K-SVD( K Singular value decomposition ) dictionary learning is proposed.First , the original signal is decomposed by VMD to obtain a series of IMF components ; second , the optimal IMF components are adaptively selected by using SAF (a spectral amplification factor )index as the training signal ; finally , the dictionary library is trained by K-SVD dictionary learning method , and the sparse signal is obtained by processing the original signal with Orthogonal Matching Pursuit ( OMP ) and is analyzed by envelope spectrum.Simulation and experimental results show that , compared with the traditional K-SVD dictionary learning method , the sparse signal-to-noise ratio ( SNR ) obtained by this method is higher , which can extract the rolling bearing periodic pulse more accurately and enhance the bearing fault characteristics.

参考文献/References:

[ 1 ] 吕明珠,苏晓明,陈长征,等 . 改进粒子群算法优化的支持向量机在滚动轴承故障诊断中的应用[ J ] . 机械与电子,2019 , 37 ( 1 ): 44-50.

[ 2 ] QINC R , WANG D D , XU Z , et al.Improved empirical wavelet transform for compound weak bearing fault diagnosis with acoustic signals [ J ] .Applied sciences , 2020 , 10 ( 2 ): 682-1-682-16.
[ 3 ] 杨蕊,李宏坤,贺长波,等 . 利用最优小波尺度循环谱的滚动轴承早期故障特征提取[ J ] . 机械工程学报, 2018 ,54 ( 17 ): 208-217.
[ 4 ] WANG J , DU G F , ZHU Z K , et al.Fault diagnosis of rotating machines based on the EMD manifold [ J ] .Mechanical systems and signal processing , 2020 , 135 ( 1 ):106443-1-106443-21.
[ 5 ] 李政,张炜,明安波,等 . 基于 IEWT 和 MCKD 的滚动轴承故障诊断方法[ J ] . 机械工程学报, 2019 , 55 ( 23 ): 136-146.
[ 6 ] 张峻宁,张培林,张云强,等 . 融合字典学习的滑动轴承摩擦故障趋势自记忆预测[?J ] . 振动工程学报, 2017 , 30(?3 ): 519-526.
[ 7 ] KONSTANTIN D , DOMINIQUE Z.Variational mode decomposition [ J ] .IEEE Transactions on signal processing , 2014 , 62 ( 3 ):531-544.
[ 8 ] ZHENG K , LI TL , SU Z Q , et al.Faults diagnosis of rolling bearings based on shift invariant K-singular value decomposition with sensitive atom nonlocal means enhancement [ J ] .Measurement , 2019 , 135 : 836-851.
[ 9 ] LIANG K X , ZHAO M , LIN J , et al.An information based K-singular value decomposition method for rolling element bearing diagnosis [ J ] .ISA Transactions , 2020 , 96 :444-456.
[ 10 ] 余发军,周凤星,严保康 . 基于字典学习的轴承早期故障稀疏特征提取[ J ] . 振动与冲击, 2016 , 35 ( 6 ): 181-186.
[ 11 ] 张文颢,李永健,张卫华 . 基于 K 奇异值分解和层次化分块正交匹配算法的滚动轴承故障诊断[ J ] . 中国机械工程,2019 , 30 ( 4 ): 406-412.
[ 12 ] LI X , MA Z Q , KANG D , et al.Fault diagnosis for rolling bearing based on VMD-FRFT [ J ] .Measurement , 2020 , 155 : 107554-1-107554-17.
[ 13 ] 王恒迪,邓四二,杨建玺,等 . 基于参数优化变分模态 分解的 滚 动 轴 承 早 期 故 障 诊 断 [ J ] . 振 动 与 冲 击,
2020 , 39( 23 ): 38-46.
[ 14 ] YANG J H , HUANG D W , ZHOU D J , et al.Optimal IMF selection and unknown fault feature extraction for rolling bearings with different defect modes [ J ] .Measurement , 2020 , 157 : 107660-1-107660-15.

备注/Memo

备注/Memo:
收稿日期: 2021-09-17
基金项目:陕西省自然科学基础研究计划( 2019JQ-852 )
作者简介:梁雄鹤 ( 1996- ),男,陕西华阴人,硕士研究生,研究方向为旋转机械故障诊断;陈 珊 ( 1988- ),女,陕西潼关人,工程师,研究方向为信号处理、数据挖掘。
更新日期/Last Update: 2022-03-02