[1]郝伟,林辉翼,郝旺身,等.基于全矢稀疏编码的滚动轴承故障识别方法[J].郑州大学学报(工学版),2019,40(03):6.[doi:10.13705/j.issn.1671-6833.2018.03.007]
 Hao Wei,Lin Huiyi,Hao Wangshen,et al.Fault Recognition Method of Rolling Bearing Based on Full Vector Sparse Coding[J].Journal of Zhengzhou University (Engineering Science),2019,40(03):6.[doi:10.13705/j.issn.1671-6833.2018.03.007]
点击复制

基于全矢稀疏编码的滚动轴承故障识别方法()
分享到:

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
40卷
期数:
2019年03期
页码:
6
栏目:
出版日期:
2019-04-30

文章信息/Info

Title:
Fault Recognition Method of Rolling Bearing Based on Full Vector Sparse Coding
作者:
郝伟林辉翼郝旺身高亚娟董辛旻
郑州大学机械工程学院
Author(s):
Hao WeiLin HuiyiHao WangshenGao YajuanDong Xinmin
School of Mechanical Engineering, Zhengzhou University
关键词:
全矢谱稀疏编码故障诊断滚动轴承字典学习
Keywords:
full spectrumsparse codingfault diagnosisRolling bearingsdictionary learning
DOI:
10.13705/j.issn.1671-6833.2018.03.007
文献标志码:
A
摘要:
针对利用时域信号进行稀疏编码存在的特征时移现象以及单通道信号分析易造成信息遗漏等问题,将全矢谱技术与稀疏编码相结合,提出了一种新的滚动轴承故障识别方法。首先对各状态下的滚动轴承同源双通道信号进行全矢信息融合;然后将融合后得到的主振矢信号进行字典学习,以构造各类信号的冗余字典;最后利用各类字典分别重构测试样本,将其重构误差的大小作为判断样本状态类别的依据。该方法直接利用全矢融合后的主振矢信号进行训练,其训练样本所包含的信息更加全面、准确,且免去了特征提取步骤,减少了人为因素的影响。实验结果表明,该方法计算效率高,稳定性好,可有效判断出滚动轴承的故障类型。
Abstract:
Aiming at the feature shift phenomenon when using time domain signals to sparse coding and the problem of single channel analysis often result in inadequate use of information, a new method for fault identification of rolling bearings is proposed by combining the full vector spectrum technique and sparse coding. Firstly, full vector information fusion of the homogeneous dual-channel signal of the rolling bearing in each state is carried out. Then, the main vibration vector signals obtained is used to construct all kinds of redundant dictionaries. Finally, these dictionaries is used to reconstruct the test samples, and the error is used as the basis for identify the status of these samples. The method use the main vibration vector signals obtained by full vector fusion as training samples, so the training samples contain more comprehensive and accurate information. Besides, the feature extraction step can be eliminated, so it can reduce the effect of human factors. The experimental results show that the method has high efficiency and good stability, and can effectively identify the fault pattern of rolling bearing.
更新日期/Last Update: 2019-04-16