Suppression of magnetotelluric square wave noise based on a LSTM recurrent neural network

YANG Kai; TANG Wei-Dong; LIU Cheng; HE Jing-Long; YAO Chuan

doi:10.11720/wtyht.2022.1572

2022 Vol. 46, No. 4

Article Contents

Article navigation > Geophysical and Geochemical Exploration > 2022 > 46(4): 925-933

Next Article Previous Article

YANG Kai, TANG Wei-Dong, LIU Cheng, HE Jing-Long, YAO Chuan. 2022. Suppression of magnetotelluric square wave noise based on a LSTM recurrent neural network. Geophysical and Geochemical Exploration, 46(4): 925-933. doi: 10.11720/wtyht.2022.1572

Citation:

YANG Kai, TANG Wei-Dong, LIU Cheng, HE Jing-Long, YAO Chuan. 2022. Suppression of magnetotelluric square wave noise based on a LSTM recurrent neural network. Geophysical and Geochemical Exploration, 46(4): 925-933. doi: 10.11720/wtyht.2022.1572

Suppression of magnetotelluric square wave noise based on a LSTM recurrent neural network

1. Xi'an Center of Mineral Resources Survey, China Geological Survey, Xi'an 710000, China；
2. Institute of Geophysics & Geomatics, China University of Geosciences (Wuhan), Wuhan 430074, China

Received Date: 25 October 2021

Revised Date: 20 August 2022

Publish Date: 17 August 2022

Abstract

Abstract

Denoising is an important part of magnetotelluric data processing. To enrich and develop the denoising method of magnetotelluric time series, this study introduced the LSTM network-one of the recurrent neural networks-into the square wave noise processing of the magnetotelluric time series. Different from previous studies, the measured magnetotelluric time series without human interference superimposed on simulated square wave noise were used as the input of the LSTM network, and the noise-free original time series were used as the target output of the network. After training for 1,500 epochs, the normalized cross-correlation coefficient between the time series extracted from the simulated noise signals by the network and the original time series reached 0.9718, indicating that the network has effectively learned the characteristics of the noise-free magnetotelluric time series. Finally, the denoising test results of measured square wave noise signals show that the proposed method can effectively suppress the interference of square wave noise and improve the estimation quality of impedance. This study provides a new idea for the processing of magnetotelluric time series based on deep learning.
- magnetotellurics /
- time series /
- LSTM /
- deep learning /
- denoising

FullText(HTML)

References

[1]	孙洁, 晋光文. 大地电磁测深资料的噪声干扰[J]. 物探与化探, 2000, 24(2):119-127. Google Scholar
[2]	Sun J, Jin G W. Noise interference of magnetotelluric data[J]. Geophysical and Ceochemical Exploration, 2000, 24(2):119-127. Google Scholar
[3]	汤井田, 徐志敏, 肖晓, 等. 庐枞矿集区大地电磁测深强噪声的影响规律[J]. 地球物理学报, 2012, 55(12):4147-4159. Google Scholar
[4]	Tang J T, Xu Z M, Xiao X, et al. Effect rules of strong noise on magnetotelluric (MT) sounding in Luzong ore cluster area[J]. Chinese Journal of Geophysics, 2012, 55(12):4147-4159. Google Scholar
[5]	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. Google Scholar
[6]	Yan J B. The study on theory and method of magnetotelluric signal processing[D]. Changsha: Central South University, 2003. Google Scholar
[7]	蔡建华, 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. Google Scholar
[8]	Cai J H, Tang J T, Wang X C. 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. Google Scholar
[9]	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. Google Scholar
[10]	Tang J T, Li J, Xiao X, et al. Mathematical morphology filtering and noise suppression of magnetotelluric sounding data[J]. Chinese Journal of Geophysics, 2012, 55(5):1784-1793. Google Scholar
[11]	汤井田, 刘祥, 周聪. 仿真方波的大地电磁远参考去噪研究[J]. 物探化探计算技术, 2014, 36(5):513-520. Google Scholar
[12]	Tang J T, Liu X, Zhou C. Simulation of square waveform de-noising research of magnetotelluric with a remote reference[J]. Computing Techniques for Geophysical and Geochemical Exploration, 2014, 36(5):513-520. Google Scholar
[13]	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. Google Scholar
[14]	Wang H, Wei W B, Jin S, et al. Removal of magnetotelluric noise based on synchronous time series relationship[J]. Chinese Journal of Geophysics, 2014, 57(2):531-545. Google Scholar
[15]	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. Google Scholar
[16]	Tang J T, Li G, Zhou C, et al. Denoising AMT data based on dictionary learning[J]. Chinese Journal of Geophysics, 2018, 61(9):3835-3850. Google Scholar
[17]	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. Google Scholar
[18]	Li J, Zhang X, Cai J. Suppression of strong interference for AMT using VMD and MP[J]. Chinese Journal of Geophysics, 2019, 62(10):3866-3884. Google Scholar
[19]	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. Google Scholar
[20]	Wang H, Yan J Y, Fu G M, et al. Current status and application prospect of deep learning in geophysics[J]. Progress in Geophysics, 2020, 35(2):642-655. Google Scholar
[21]	李瑞友, 张淮清, 吴昭. 基于在线惯序极限学习机的瞬变电磁非线性反演[J]. 物探与化探, 2021, 45(4):1048-1054. Google Scholar
[22]	Li R Y, Zhang H Q, Wu Z. Online sequential extreme learning machine for transient electromagnetic nonlinear inversion[J]. Geophysical and Geochemical Exploration, 2021, 45(4):1048-1054. Google Scholar
[23]	王逸宸, 柳林涛, 许厚泽. 基于卷积神经网络识别重力异常体[J]. 物探与化探, 2020, 44(2):394-400. Google Scholar
[24]	Wang Y C, Liu L T, Xu H Z. The identification of gravity anomaly body based on the convolutional neural network[J]. Geophysical and Geochemical Exploration, 2020, 44(2):394-400. Google Scholar
[25]	韩卫雪, 周亚同, 池越. 基于深度学习卷积神经网络的地震数据随机噪声去除[J]. 石油物探, 2018, 57(6):862-869. Google Scholar
[26]	Han W X, Zhou Y T, Chi Y. Deep learning convolutional neural networks for random noise attenuation in seismic data[J]. Geophysical Prospecting for Petroleum, 2018, 57(6):862-869. Google Scholar
[27]	王鹤, 蒋欢, 王亮, 等. 大地电磁人工神经网络反演[J]. 中南大学学报:自然科学版, 2015, 46(5):1707-1714. Google Scholar
[28]	Wang H, Jiang H, Wang L, et al. Magnetotelluric inversion using artificial neural network[J]. Journal of Central South University:Science and Technology, 2015, 46(5):1707-1714. Google Scholar
[29]	赵文举, 刘云祥, 陶德强, 等. BP神经网络磁性体顶面埋深预测方法[J]. 石油地球物理勘探, 2020, 55(5):1139-1148. Google Scholar
[30]	Zhao W J, Liu Y X, Tao D Q, et al. Prediction of magnetic body top based on BP neural network[J]. Oil Geophysical Prospecting, 2020, 55(5):1139-1148. Google Scholar
[31]	Manoj C, Nagarajan N. The application of artificial neural networks to magnetotelluric time-series analysis[J]. Geophysical Journal International, 2003, 152(2):409-423. Google Scholar
[32]	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. Google Scholar
[33]	Xu T T, Wang Z X, Xiao Z W, et al. Magnetotelluric power frequency interference suppression based on LSTM recurrent neural network[J]. Progress in Geophysics, 2020, 35(5):2016-2022. Google Scholar
[34]	王斯昊, 何兰芳. 基于LSTM循环神经网络的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, 2021. Google Scholar
[35]	Wang S H, He N F. Feasibility analysis of the LSTM based denoising to MT time series[C]// The 9^th Conference of Geophysical Technology,Chinese Geophysical Society—Abstracts of Symposium on Full-Space Geophysical Exploration and Intelligent Sensing, 2021. Google Scholar
[36]	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. Google Scholar
[37]	Wang K X, Huang Q H, Wu S H. Application of long short-term memory neural network in geoelectric field data processing[J]. Chinese Journal of Geophysics, 2020, 63(8):3015-3024. Google Scholar
[38]	Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780. Google Scholar
[39]	Kingma D, Ba J. Adam:A method for stochastic optimization[J]. Computer Science, 2014(12).doi: 10.48550/arXiv.1412.6980. Google Scholar
[40]	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. Google Scholar
[41]	Yang Y, Wang X, Han J, et al. Magnetotelluric transfer function distortion assessment using Nyquist diagrams[J]. Journal of Applied Geophysics, 2019, 160:218-228. Google Scholar
[42]	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. Google Scholar
[43]	Wang S M, Wang J Y. Analysis on statistic characteristics of Magnetotelluric signal[J]. Acta Seismologica Sinica, 2004, 26(6):669-674. Google Scholar
[44]	Neukirch M, Garcia X. Nonstationary magnetotelluric data processing with instantaneous parameter[J]. Journal of Geophysical Research:Solid Earth, 2014, 119(3):1634-1654. Google Scholar