Processing of impulse noise in magnetotelluric data based on bidirectional recurrent imputation

Authors

YANG Kai, Xi’an Center of Mineral Resources Survey, China Geological Survey, Xi’an 710100, China; Technology Innovation Center for Gold Ore Exploration, China Geological Survey, Xi’an 710100, China; School of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, ChinaFollow
LIU Cheng, Technology Innovation Center for Gold Ore Exploration, China Geological Survey, Xi’an 710100, China; School of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, China
LI Han, Technology Innovation Center for Gold Ore Exploration, China Geological Survey, Xi’an 710100, China; School of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, China
HE Jinglong, Technology Innovation Center for Gold Ore Exploration, China Geological Survey, Xi’an 710100, China; School of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, China

Abstract

Objective Magnetotelluric sounding is an exploration method to obtain the underground electrical structure by observing the natural electromagnetic field, which is easily disturbed by noise. Impulse noise, frequently occurring in magnetotelluric sounding, generally exhibits high amplitude and wide frequency bands, producing significant impacts on the data quality. Methods To suppress such noise, this study proposed a method based on a bidirectional recurrent imputation for time series (BRITS) model. First, the data segments with noise interference were deleted. Second, for the magnetotelluric time series with missing data to be imputed, training sets were constructed for imputation training of the BRITS model. Third, imputation was conducted to supplement the missing data, yielding the denoising results. Last, the proposed method was applied to process the simulated and measured data with noise, and the application results were compared with the results derived using the empirical mode decomposition (EMD) threshold method. Results and Conclusions The results of this study are as follows: (1) Relative to the original data, the simulated data with noise, after being processed using the BRITS method, manifested normalized cross-coefficient reaching up to 0.999 and signal-to-noise ratios of over 29 dB. In contrast, the simulated noise data, after being treated using the EMD threshold method displayed cross-coefficient of 0.778 and signal-to-noise of 3.09 dB. (2) In the processing of the measured data, the BRITS method effectively restored the data with noise interference, with the obtained Nyquist diagram closer to the characteristics of natural magnetotelluric signals compared to the EMD threshold method. (3) As indicated by the test results of different training samples, in the case where four-component magnetotelluric data contain at least two normal components, with the proportion of noise in a single unnormal component not exceeding 20% and the continuous noise interference length of 10 sampling points or less, the data denoised using the BRITS method can yield cross-coefficient exceeding 0.96, thus ensuring certain denoising accuracy.

Keywords

magnetotelluric sounding, noise removal, impulse noise, time series imputation, bidirectional recurrent imputation

DOI

10.12363/issn.1001-1986.24.01.0084

Recommended Citation

YANG Kai, LIU Cheng, LI Han, et al. (2024) "Processing of impulse noise in magnetotelluric data based on bidirectional recurrent imputation," Coal Geology & Exploration: Vol. 52: Iss. 8, Article 19.
DOI: 10.12363/issn.1001-1986.24.01.0084
Available at: https://cge.researchcommons.org/journal/vol52/iss8/19

Reference

[1] 严家斌. 大地电磁信号处理理论及方法研究[D]. 长沙：中南大学，2003.

YAN Jiabin. The study on theory and method of magnetotelluric signal processing[D]. Changsha：Central South University，2003.

[2] 严家斌，刘贵忠. 基于小波变换的脉冲类电磁噪声处理[J]. 煤田地质与勘探，2007，35(5)：61−65.

YAN Jiabin，LIU Guizhong. Like-impulse electromagnetic noise processing based wavelet transform[J]. Coal Geology & Exploration，2007，35(5)：61−65.

[3] 蔡剑华，汤井田，王先春. 基于经验模态分解的大地电磁资料人文噪声处理[J]. 中南大学学报(自然科学版)，2011，42(6)：1786−1790.

CAI Jianhua，TANG Jingtian，WANG Xianchun. Human noise elimination for magnetotelluric data based on empirical mode decomposition[J]. Journal of Central South University (Science and Technology)，2011，42(6)：1786−1790.

[4] 王正海，耿欣，姚卓森，等. 基于经验模态分解的MT数据脉冲类电磁噪声处理[J]. 吉林大学学报(地球科学版)，2014，44(1)：344−349.

WANG Zhenghai，GENG Xin，YAO Zhuosen，et al. Like-impulse electromagnetic noise processing based on empirical mode decomposition[J]. Journal of Jilin University (Earth Science Edition)，2014，44(1)：344−349.

[5] 汤井田，刘子杰，刘峰屹，等. 音频大地电磁法强干扰压制试验研究[J]. 地球物理学报，2015，58(12)：4636−4647.

TANG Jingtian，LIU Zijie，LIU Fengyi，et al. The denoising of the audio magnetotelluric data set with strong interferences[J]. Chinese Journal of Geophysics，2015，58(12)：4636−4647.

[6] 汤井田，李晋，肖晓，等. 数学形态滤波与大地电磁噪声压制[J]. 地球物理学报，2012，55(5)：1784−1793.

TANG Jingtian，LI Jin，XIAO Xiao，et al. Mathematical morphology filtering and noise suppression of magnetotelluric sounding data[J]. Chinese Journal of Geophysics，2012，55(5)：1784−1793.

[7] 汤井田，李晋，肖晓，等. 基于数学形态滤波的大地电磁强干扰分离方法[J]. 中南大学学报(自然科学版)，2012，43(6)：2215−2221.

TANG Jingtian，LI Jin，XIAO Xiao，et al. Magnetotelluric sounding data strong interference separation method based on mathematical morphology filtering[J]. Journal of Central South University (Science and Technology)，2012，43(6)：2215−2221.

[8] 汤磊，汤洪志，韩雪平，等. 数学形态滤波在音频大地电磁去噪中的应用[J]. 工程地球物理学报，2013，10(4)：533−538.

TANG Lei，TANG Hongzhi，HAN Xueping，et al. The application of mathematical morphological filtering to the noise suppression of audio magnetotelluric sounding data[J]. Chinese Journal of Engineering Geophysics，2013，10(4)：533−538.

[9] 李晋，张贤，蔡锦. 利用变分模态分解(VMD)和匹配追踪(MP)联合压制音频大地电磁(AMT)强干扰[J]. 地球物理学报，2019，62(10)：3866−3884.

LI Jin，ZHANG Xian，CAI Jin. Suppression of strong interference for AMT using VMD and MP[J]. Chinese Journal of Geophysics，2019，62(10)：3866−3884.

[10] 李晋，燕欢，汤井田，等. 基于匹配追踪和遗传算法的大地电磁噪声压制[J]. 地球物理学报，2018，61(7)：3086−3101.

LI Jin，YAN Huan，TANG Jingtian，et al. Magnetotelluric noise suppression based on matching pursuit and genetic algorithm[J]. Chinese Journal of Geophysics，2018，61(7)：3086−3101.

[11] 舒幸宁. 基于AR模型的大地电磁脉冲噪声压制[J]. 科技创新与应用，2021，11(18)：102−104.

SHU Xingning. Suppression of magnetotelluric pulse noise based on AR model[J]. Technology Innovation and Application，2021，11(18)：102−104.

[12] 汤井田，李广，周聪，等. 基于字典学习的音频大地电磁数据处理[J]. 地球物理学报，2018，61(9)：3835−3850.

TANG Jingtian，LI Guang，ZHOU Cong，et al. Denoising AMT data based on dictionary learning[J]. Chinese Journal of Geophysics，2018，61(9)：3835−3850.

[13] 桂团福，邓居智，李广，等. 数学形态学和K-SVD字典学习在大地电磁数据去噪中的应用[J]. 中国有色金属学报，2021，31(12)：3713−3729.

GUI Tuanfu，DENG Juzhi，LI Guang，et al. De-noising magnetotelluric data based on mathematical morphology and K-SVD dictionary learning[J]. The Chinese Journal of Nonferrous Metals，2021，31(12)：3713−3729.

[14] 王昊，严加永，付光明，等. 深度学习在地球物理中的应用现状与前景[J]. 地球物理学进展，2020，35(2)：642−655.

WANG Hao，YAN Jiayong，FU Guangming，et al. Current status and application prospect of deep learning in geophysics[J]. Progress in Geophysics，2020，35(2)：642−655.

[15] 韩盈，安志国，屈文璋. 基于机器学习的大地电磁时域数据处理的研究现状[J]. 地球物理学进展，2021，36(5)：1975−1987.

HAN Ying，AN Zhiguo，QU Wenzhang. Research status of magnetotelluric time domain data processing based on machine learning[J]. Progress in Geophysics，2021，36(5)：1975−1987.

[16] YU Siwei，MA Jianwei. Deep learning for geophysics：Current and future trends[J]. Reviews of Geophysics，2021，59(3)：e2021RG000742.

[17] 许滔滔，王中兴，肖卓伟，等. 基于LSTM循环神经网络的大地电磁工频干扰压制[J]. 地球物理学进展，2020，35(5)：2016−2022.

XU Taotao，WANG Zhongxing，XIAO Zhuowei，et al. Magnetotelluric power frequency interference suppression based on LSTM recurrent neural network[J]. Progress in Geophysics，2020，35(5)：2016−2022.

[18] 王斯昊，何兰芳. 基于LSTM循环网络的MT时间序列去噪可行性分析[C]//中国地球物理学会，地球物理技术委员会. 第九届学术会议—全域地球物理探测与智能感知学术研讨会会议摘要集. 武汉，2021：355–357.

[19] 汪凯翔，黄清华，吴思弘. 长短时记忆神经网络在地电场数据处理中的应用[J]. 地球物理学报，2020，63(8)：3015−3024.

WANG Kaixiang，HUANG Qinghua，WU Sihong. Application of long short-term memory neural network in geoelectric field data processing[J]. Chinese Journal of Geophysics，2020，63(8)：3015−3024.

[20] 杨凯，唐卫东，刘诚，等. 基于LSTM循环神经网络的大地电磁方波噪声抑制[J]. 物探与化探，2022，46(4)：925−933.

YANG Kai，TANG Weidong，LIU Cheng，et al. Suppression of magnetotelluric square wave noise based on a LSTM recurrent neural network[J]. Geophysical and Geochemical Exploration，2022，46(4)：925−933.

[21] CAO Wei，WANG Dong，LI Jian，et al. BRITS：bidirectional recurrent imputation for time series[C]//Proceedings of the 32nd International Conference on Neural Information Processing Systems. Cambridge：MIT Press，2018：6776–6786.

[22] XIE Chuanjie，HUANG Chong，ZHANG Deqiang，et al. BiLSTM-I：A deep learning-based long interval gap-filling method for meteorological observation data[J]. International Journal of Environmental Research and Public Health，2021，18(19)：10321.

[23] 郑欣彤，边婷婷，张德强，等. ARIMA和LSTM方法长时间温度观测数据缺失值插补的比较[J]. 计算机应用，2022，42(增刊1)：130−135.

ZHENG Xintong，BIAN Tingting，ZHANG Deqiang，et al. Comparison of ARIMA and LSTM methods for interpolation of missing values of long-time temperature observations[J]. Journal of Computer Applications，2022，42(Sup.1)：130−135.

[24] 朱政. 雷达高分辨距离像缺失信号重构与目标识别方法研究[D]. 西安：西安电子科技大学，2022.

ZHU Zheng. Research on reconstruction and target recognition of missing signal of radar high resolution range profile[D]. Xi’an：Xidian University，2022.

[25] 严家斌，刘贵忠. 天然大地电磁场时间序列的分形特征[J]. 煤田地质与勘探，2007，35(2)：66−69.

YAN Jiabin，LIU Guizhong. Fractal characteristic of time series of nature field[J]. Coal Geology & Exploration，2007，35(2)：66−69.

[26] 王辉，魏文博，金胜，等. 基于同步大地电磁时间序列依赖关系的噪声处理[J]. 地球物理学报，2014，57(2)：531−545.

WANG Hui，WEI Wenbo，JIN Sheng，et al. Removal of magnetotelluric noise based on synchronous time series relationship[J]. Chinese Journal of Geophysics，2014，57(2)：531−545.

[27] WANG Hui，CAMPANYÀ J，CHENG Jiulong，et al. Synthesis of natural electric and magnetic Time-series using Inter-station transfer functions and time-series from a Neighboring site (STIN)：Applications for processing MT data[J]. Journal of Geophysical Research：Solid Earth，2017，122(8)：5835−5851.

[28] 王书明，王家映. 大地电磁信号统计特征分析[J]. 地震学报，2004，26(6)：669−674.

WANG Shuming，WANG Jiaying. Analysis on statistic characteristics of magnetotelluric signal[J]. Acta Seismologica Sinica，2004，26(6)：669−674.

[29] NEUKIRCH M，GARCIA X. Nonstationary magnetotelluric data processing with instantaneous parameter[J]. Journal of Geophysical Research：Solid Earth，2014，119(3)：1634−1654.

[30] UHM J，HEO J，MIN D J，et al. Imaging strategies to interpret 3-D noisy audio-magnetotelluric data acquired in Gyeongju，South Korea：Data processing and inversion[J]. Geophysical Journal International，2021，225(2)：744−758.

[31] YANG Yue，WANG Xueqiu，HAN Jiangtao，et al. Magnetotelluric transfer function distortion assessment using Nyquist diagrams[J]. Journal of Applied Geophysics，2019，160：218−228.