基于FDM的无损检测声发射信号仿真分析

Acoustic Emission Simulation Signal Denoising Analysis based on FDM

为了把声发射信号中的无用信息(噪声信息)分离出来,可以更加精准的提取声发射信号的信息参数,对其进行后续分析,针对声发射信号的非平稳性、非周期性等特点,引入了一种傅里叶分解(FDM)的声发射(AE)信号去噪分析方法。将复杂声发射信号分解到各个频率区间内,对分解所得优势分量进行重组,对重组信号进行时频分析。傅里叶分解区别于以往常用的EMD、EEMD分解方法,可以有效的避免模态混叠现象,为声发射信号去噪及时频分析提供了新途径。

In order to separate the useless information(noise information)in acoustic emission signals,and extract the information parameters of acoustic emission signals more accurately,in view of the non-stationarity and non-periodicity of acoustic emission signals,a Fourier decomposition(FDM)de-noising analysis method for acoustic emission(AE)signals is introduced in this paper.The complex acoustic emission signals are decomposed into each frequency interval,and the dominant components obtained from the decomposition are reassembled,and the time frequency analysis of the reassembled signals is carried out.Fourier decomposition is different from EMD and EEMD decomposition methods commonly used in the past,which can effectively avoid the mode aliasing phenomenon.At the same time,no white noise is added in the process of Fourier decomposition,which can also effectively avoid the residual problem of white noise,providing a new way for acoustic emission signal denoising and frequency analysis.