基于共振稀疏分解的滚动轴承早期微弱故障诊断
发布时间：2018-07-15 12:12
【摘要】：传统方法很难对滚动轴承的早期微弱故障进行有效诊断.共振稀疏分解是一种基于多字典库的稀疏分解方法,可以同时分解出滚动轴承故障信号中的瞬态冲击成分及其持续震荡成分(工频及其谐频成分).该方法在对滚动轴承早期微弱故障信号进行自适应滤波降噪(采用Ensemble Empirical Mode Decomposition,EEMD方法)基础上,对处理后的信号进行共振稀疏分解分析,分别构建高、低品质因子小波基函数字典库,并利用形态学分析方法建立信号稀疏表示的目标函数,进而实现对滚动轴承发生故障时具有低品质因子的瞬态故障成分及其他持续振荡高品质因子噪声成分的成功分离.对分离得到的低品质因子信号成分进行包络解调分析,进而得到较好的故障提取特征结果.通过实验验证了所述方法的有效性.
[Abstract]:The traditional method is difficult to diagnose the early weak fault of rolling bearing effectively. Resonance sparse decomposition is a sparse decomposition method based on multi-dictionary library, which can simultaneously decompose the transient shock component and its sustained oscillation component (power frequency and harmonic component) in the fault signal of rolling bearing. This method is based on adaptive filtering noise reduction (Ensemble empirical Mode Decompositionation EEMD method) for the early weak fault signal of rolling bearing, and the resonance sparse decomposition analysis of the processed signal is carried out. The low quality factor wavelet basis function dictionary and the objective function of sparse signal representation are established by morphological analysis. Then the transient fault components with low quality factors and the noise components of other continuous oscillation high quality factors are successfully separated from each other when the rolling bearings fail. The envelope demodulation analysis is carried out on the components of the separated low quality factor signal, and better fault feature extraction results are obtained. The effectiveness of the method is verified by experiments.
【作者单位】： 国家电投集团河南电力有限公司;
【基金】：河南省高等学校精密制造技术与工程重点学科开放实验室开放基金资助项目(PMTE201302A)
【分类号】：TH133.33
本文编号：2124031
[Abstract]:The traditional method is difficult to diagnose the early weak fault of rolling bearing effectively. Resonance sparse decomposition is a sparse decomposition method based on multi-dictionary library, which can simultaneously decompose the transient shock component and its sustained oscillation component (power frequency and harmonic component) in the fault signal of rolling bearing. This method is based on adaptive filtering noise reduction (Ensemble empirical Mode Decompositionation EEMD method) for the early weak fault signal of rolling bearing, and the resonance sparse decomposition analysis of the processed signal is carried out. The low quality factor wavelet basis function dictionary and the objective function of sparse signal representation are established by morphological analysis. Then the transient fault components with low quality factors and the noise components of other continuous oscillation high quality factors are successfully separated from each other when the rolling bearings fail. The envelope demodulation analysis is carried out on the components of the separated low quality factor signal, and better fault feature extraction results are obtained. The effectiveness of the method is verified by experiments.
【作者单位】： 国家电投集团河南电力有限公司;
【基金】：河南省高等学校精密制造技术与工程重点学科开放实验室开放基金资助项目(PMTE201302A)
【分类号】：TH133.33
【相似文献】
相关期刊论文 前2条
1 孙云嵩;于德介;陈向民;李蓉;;基于信号共振稀疏分解的阶比分析及其在齿轮故障诊断中的应用[J];振动与冲击;2013年16期
2 张文义;于德介;陈向民;;基于信号共振稀疏分解与能量算子解调的轴承故障诊断方法[J];中国电机工程学报;2013年20期
相关博士学位论文 前1条
1 王宏超;基于稀疏分解及图像稀疏表征的滚动轴承微弱故障诊断[D];上海交通大学;2015年
相关硕士学位论文 前7条
1 张顶成;基于最优信号共振稀疏分解的旋转机械故障诊断方法研究[D];湖南大学;2016年
2 骆杰;改进的原子稀疏分解算法及其在机械故障诊断中的应用研究[D];武汉科技大学;2016年
3 龚永涛;改进的共振稀疏分解方法及在机械故障诊断中的应用[D];武汉科技大学;2016年
4 孙占龙;基于共振稀疏分解的滚动轴承故障诊断方法研究[D];北京交通大学;2017年
5 郑晓慧;机械振动信号的稀疏分解理论研究[D];兰州理工大学;2014年
6 李星;基于信号共振稀疏分解的齿轮箱故障诊断方法研究[D];湖南大学;2014年
7 孙云嵩;基于信号共振稀疏分解的齿轮故障诊断方法研究[D];湖南大学;2013年
，本文编号：2124031
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2124031.html
