井下地磁匹配数据的小波降噪性能研究  

Wavelet Denoising Performance of Underground Geomagnetic Matching Data Preprocessing

作　　者：汪金花 杨华文 李鸣铎 刘暑明 WANG Jinhua;YANG Huawen;LI Mingduo;LIU Shuming(College of Mining Engineering,North China University of Technology,Tangshan,Hebei 063210,China;College of Electrical Engineering,North China University of Technology,Tangshan,Hebei 063210,China)

机构地区：[1]华北理工大学矿业工程学院,河北唐山063210 [2]华北理工大学电气工程学院,河北唐山063210

出　　处：《昆明理工大学学报（自然科学版）》2022年第2期35-46,共12页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：国家自然科学基金面上项目(42074014);河北省自然科学基金项目(E2018209345);河北省教育厅重点研究项目(ZD2021082)。

摘　　要：井下地磁定位中地磁基准图或实测地磁序列的噪声扰动是地磁匹配定位结果不确定的重要原因之一.为了研究地磁匹配定位数学模型去除噪声的效果,选取了4个不同适配性的研究区,开展了不同阈值方法下小波去噪的灵敏性测试,进行了傅立叶滤波和小波阈值去噪试验,并选用MSD匹配定位验证了小波去噪的可靠性.试验结果表明,默认阈值、自适应阈值和极大极小阈值方法的小波去噪效果不一样,对3 500 nT高斯噪声,自适应阈值wav-R2小波去噪效果最好;FFT变换和wav-R2小波变换均有一定的降噪效果,但wav-R2小波变换局部去噪更精准,能够保持局部磁变化微小特征.对700、2 100、3 500、4 900和6 300 nT的高斯噪声采用小波自适应阈值降噪后,发现小波去噪具有良好的稳定性,去噪后的信噪比接近已知理论值.磁数据wav-R2小波去噪能够显著提高磁匹配效果,去噪后的MSD匹配试验平均概率达到了90%,为磁匹配定位降噪预处理提供了参考方法.The noise of geomagnetic reference map or measured geomagnetic sequence in underground geomagnetic positioning is one of the key factors of geomagnetic matching uncertainty.In order to study the denoising effect of geomagnetic noise reduction model,four research areas with different adaptability are selected,the sensitivity test of different threshold methods of wavelet denoising as well as Fourier filtering and wavelet threshold denoising test are carried out,and MSD matching location is selected to verify the reliability of wavelet denoising.The test results show that the wavelet denoising effects of default threshold,adaptive threshold and maximum and minimum threshold are different.For 3 500 nT Gaussian noise,the adaptive threshold wav-R2 wavelet denoising effect is the best;FFT transform and wav-R2 wavelet transform can reduce noise to some extent,but wav-R2 wavelet transform is more accurate in local denoising and retain the characteristics of local small magnetic changes.After wavelet adaptive threshold denoising for 700,2 100,3 500,4 900 and 6 300 nT Gaussian noise,wavelet denoising has good stability,and the signal-to-noise ratio after denoising is close to the known value.The magnetic data wav-R2 wavelet denoising can significantly improve the magnetic matching probability.The average probability of MSD matching after experimental denoising reaches 90%,which provides a reference method for magnetic matching location noise reduction preprocessing.

关 键 词：地磁匹配 地磁数据预处理 噪声扰动 小波阈值去噪 FFT去噪 

分 类 号：P318[天文地球—固体地球物理学]