基于改进压缩采样匹配追踪算法的机械振动信号恢复  

Mechanical Vibration Signal Repair Based on Improved Compression Sampling Matching Tracking Algorithm

作　　者：李一飞 王桂宝 李伟 王磊 杨坤 王楠[1] LI Yifei;WANG Guibao;LI Wei;WANG Lei;YANG Kun;WANG Nan(School of Mechanical Engineering,Shaanxi University of Technology,Hanzhong Shaanxi 723001,China)

机构地区：[1]陕西理工大学机械工程学院,陕西汉中723001

出　　处：《机床与液压》2024年第2期204-208,共5页Machine Tool & Hydraulics

基　　金：国家自然科学基金面上项目(61972239);陕西省重点研发计划(2020GY-024,2021GY-182);陕西交通科技项目(22-19X)。

摘　　要：为了解决压缩感知重构算法对于机械振动信号的残缺数据恢复不佳的问题,提出一种改进压缩采样匹配追踪算法,对缺失信号进行修复重构。对比几种同类的贪婪重构算法恢复缺失信号的效果。通过仿真数据和实测数据验证算法对信号的恢复效果。结果表明:改进方法能够很好地实现对缺损信号的修复,且重构概率远远高于其他重构算法,比较压缩采样匹配追踪算法与其改进算法发现:在稀疏度为50时,改进算法的重构概率可以达到100%,而未改进的压缩采样匹配追踪算法重构概率为0,说明改进算法的重构效果优于原算法,重构出来的信号可以准确地表现原始信号的全部信息。In order to solve the problem of poor data recovery of mechanical vibration signals by compressed sensing reconstruction algorithm,an improved compressed sampling matching tracking algorithm was proposed to repair and reconstruct the missing signals.The effect of several similar greedy reconstruction algorithms in recovering missing signals was compared.The recovery effect of the algorithm on the signal was verified by simulation data and measured data.The results show that the improved method can well realize the repair of the defective signal,and the reconstruction probability is much higher than that of other reconstruction algorithms,and the comparison of the compressed sampling matching tracking algorithm and the improved algorithm show that when the sparsity is 50,the reconstruction probability of the improved algorithm can reach 100%,while the reconstruction probability of the unimproved compressed sampling matching tracking algorithm is 0,which indicates that the reconstruction effect of the improved algorithm is better than that of the origi⁃nal algorithm,and the reconstructed signal can accurately represent all the information of the original signal.

关 键 词：振动信号修复 压缩感知 重构效果 改进压缩采样匹配追踪算法 

分 类 号：TN98[电子电信—信息与通信工程] TH393[机械工程—机械制造及自动化]