局部奇异值分解降噪的研究  

Noise Reduction Based on Partial Singular Value Decomposition

作　　者：龚木红 高永毅[2] 李赞[1] 袁继广 

机构地区：[1]湖南科技大学机械设备健康维护湖南省重点实验室,湖南湘潭411201 [2]湖南科技大学物理与电子科学学院,湖南湘潭411201

出　　处：《机械工程师》2016年第5期76-78,共3页Mechanical Engineer

基　　金：国家自然科学基金(51175172);国家自然科学基金(51274098);国家自然科学基金(51174088);国家自然科学基金(51134005)

摘　　要：利用测试信号构造连续截断矩阵进行奇异值分解(SVD)是消除随机噪声干扰的有效方法,针对时域信号较长在分解中无法自适应确定分离阶数的问题,提出一种分段串联奇异值分解降噪方法。该方法先将信号连续截断成分段信号,然后分别进行奇异值分解的降噪,再通过拼接得到降噪后的信号。仿真及实验结果表明,该方法有效。利用该方法处理混凝土泵车的振动噪声测试信号,可有效提高其信噪比,最大程度地优化信号去噪的效果,提高分析的可靠性。Using the test signal to construct a continuous truncated matrix which using for the singular value decomposition（SVD） is an effective method to eliminate the random noise. Time domain signal with noise is long. It can＇t determine the separate order number. So this paper proposes a method that segments the signal firstly and then uses SVD. It can splice the segments to reduction signal using this method. The noise and vibration signal measured on concrete pump truck and mechanical equipment could be refined through this method to improve the signal to noise radio,optimize the noise elimination effect as much as possible, and the analysis which depends on the signal is more reliable.

关 键 词：奇异值分解 曲率谱 混凝土泵车 信噪比 振动噪声 

分 类 号：TH39[机械工程—机械制造及自动化]