摘要
为了解决工程类变形监测数据受噪声干扰,无法表现变形体真实变形趋势的问题,本文在传统局部均值分解(LMD)技术的基础上引入小波包去噪法,构建新的基于LMD的小波包去噪法.该组合去噪法实现去噪的主要步骤是:首先,使用LMD对原始信号进行分解,得到若干个分量,其中,信号的有用信息主要包含在低频分量中,噪声成分包含在高频分量中;其次,使用小波包去噪法对高频分量进行去噪,尽可能提取高频分量中的有用信息;最后,将去噪处理的高频分量与低频分量进行重构,得到最终去噪信号.另外,还使用仿真信号与实测变形监测信号对本文提出的去噪方法进行了检验.结果表明,相较于传统的LMD去噪法与小波包去噪法,本文基于LMD的小波包去噪法得到的相关系数、信噪比更高,均方根误差更低,具有更好的去噪效果.
Abstract
The wavelet packet denoising method was introduced in this paper based on traditional LMD technology to construct a new LMD based wavelet packet denoising method in order to solve the problems of engineering deformation monitoring data being disturbed by noise and unable to represent the true deformation trend of deformation bodies.The main steps used by this combined denoising method are as following.Firstly,LMD is used to decompose the original signal into several components,where the useful information of the signal is mainly contained in the low-frequency,and the noise is contained in the high-frequency.Secondly,wavelet packet method is used to denoise high-frequency and extract useful informa-tion from them as much as possible.Finally,both the high-frequency and low-frequency components of the denoising process are reconstructed to obtain the final denoised signal.Moreover,the denoising method proposed in this paper is validated using simulated signals and measured deformation monitoring signals.The results show that,compared with traditional LMD denoising methods and wavelet packet denoising methods,higher correlation coefficients and signal-to-noise ratio,lower root mean square error,and better denoising effects can be obtained by the denoising method proposed in this paper.
NETL
NSTL
维普
万方数据