时域压缩特征提取及压缩感知在设备状态评估中的应用研究

发布时间：2018-06-04 07:03

本文选题：压缩感知 + 滚动轴承　；参考：《机械科学与技术》2017年10期

【摘要】：压缩感知是一种新的信号采集与处理框架,其框架中压缩采样过程能够直接获取"压缩"的采样数据。本文中研究了如何利用这些压缩数据提取特征并用于设备的状态评估。首先在压缩感知框架下研究压缩采样数据的特点,研究压缩数据的压缩性与信号的稀疏性的对应关系;接着提出一种时域压缩特征计算方法,用于提取压缩数据的特征信息;最后以滚动轴承为对象,使用时域压缩特征对滚动轴承的运行状态进行评估。使用滚动轴承全寿命周期数据进行实验分析,实验结果表明,时域压缩特征能够准确的判断轴承的运行状态。
[Abstract]:Compression sensing is a new frame for signal acquisition and processing, in which the compressed sampling process can directly obtain "compressed" sampling data. In this paper, we study how to use these compressed data to extract features and to evaluate the state of equipment. Firstly, the characteristics of compressed sampled data are studied in the frame of compression perception, and the corresponding relation between compressed data compressibility and signal sparsity is studied, and then a time-domain compression feature calculation method is proposed to extract the feature information of compressed data. Finally, the rolling bearing is evaluated by time domain compression feature. The experimental results show that the time domain compression feature can accurately judge the running state of the rolling bearing by using the data of the whole life cycle of the rolling bearing.
【作者单位】：昆明云内动力股份有限公司;昆明理工大学机电工程学院;
【基金】：国家自然科学基金项目(51405211) 云南省教育厅科学研究基金项目(2013Y311)资助
【分类号】：TH133.33;TN911.7

【相似文献】