基于自寻优小波降噪算法的海洋重力数据滤波  

Marine gravity data filtering based on self-optimizing wavelet denoising algorithm

作　　者：李冬毅 覃方君[1] 黄春福 李安[1] LI Dongyi;QIN Fangjun;HUANG Chunfu;LI An(School of Electrical Engineering,Naval University of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学电气工程学院,武汉430033

出　　处：《中国惯性技术学报》2023年第9期883-889,共7页Journal of Chinese Inertial Technology

基　　金：国家自然科学基金(61873275,42274013)。

摘　　要：为抑制船测重力数据中的噪声,提升海洋重力场数据精度,提出了可自动修正阈值参数的自寻优小波降噪算法。首先比较了IIR滤波器、FIR滤波器、Kalman滤波方法、传统小波降噪方法的滤波效果,分析了几种方法的特点及适用性。其次,根据海洋重力信号大多属于缓变的特点,借鉴遗传算法思想,设计了自寻优小波降噪算法。最后,利用湖上船载原子重力仪测量数据进行了实验验证,实验结果表明,自寻优小波降噪算法解决了传统小波降噪方法泛化能力弱、对复杂噪声分量滤除不彻底的问题,与IIR滤波器、FIR滤波器、Kalman滤波方法和传统小波降噪方法等滤波算法相比,滤波结果更接近有用信号,不需要对数据进行截短,滤波后信号精度提升了17%以上。In order to suppress the noise in ship gravity data and improve the accuracy of Marine gravity field data,a self-optimizing wavelet denoising algorithm which can automatically correct threshold parameters is proposed.Fristly,the filtering effects of IIR filter,FIR filter,Kalman filter and traditional wavelet denoising method are compared,and the characteristics and applicability of these methods are analyzed.Secondly,according to that most of the Marine gravity signals are slow changing,a self-optimizing wavelet denoising algorithm is designed for reference to the genetic algorithm.Finally,experimental verification is carried out using the atomic gravimeter measurement data on the lake ship.The experimental results show that the self-optimizing wavelet denoising algorithm solves the problem that the traditional wavelet denoising method has weak generalization ability and incomplete filtering of complex noise components.Compared with the filtering algorithms such as IIR filter,FIR filter,Kalman filter and traditional wavelet denoising method,the filtering results are closer to useful signals,and there is no need to truncate the data.After filtering,the signal accuracy is improved by more than 17%.

关 键 词：海洋重力测量 信号处理 小波降噪 遗传算法 

分 类 号：U666.1[交通运输工程—船舶及航道工程]