基于人工神经网络的大地电磁时序分类研究

doi:10.11720/wtyht.2024.1144

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (6330 KB) HTML (0 KB)
Export: BibTeX | EndNote (RIS)

Abstract

With the development of society,high-quality magnetotelluric signal acquisition is becoming more and more difficult because of various types of human interference are increasingly intensified. Scholars have proposed many corresponding denoising methods for different types of noise to improve data quality. It is impossible to manually interpret each data before denoising due to the huge amount of data. So an efficient noise recognition and classification method is urgently needed. Based on this, artificial neural network is applied in the classification of magnetotelluric time series in this paper. Four types of time series,namely,simulated square wave, power frequency, impulse noise, and measured noiseless data, were used to conduct noise classification training and measured data classification on LSTM、FCN、ResNet、LSTM-FCN and LSTM-ResNet models. The results show that FCN and LSTM-FCN has a relatively good effect on the classification of magnetotelluric time series. Among them, the highest classification accuracy of FCN measured data can reach 99.84%, and the average time for each epoch is 9.6 s. LSTM-FCN has higher classification accuracy than FCN,the highest classification accuracy of measured data sets is nearly 100%, but the average time for each epoch is 24.6 s, and it is easier to overfit than FCN. Overall, LSTM-FCN can achieve higher classification accuracy when the amount of data is relatively small, if the amount of data is large, it is necessary to consider the time cost,using FCN is more appropriate. Finally,the magnetotelluric data containing different types of noise was successfully processed using the magnetotelluric noise processing system which constructed by the LSTM-FCN classification model and the LSTM denoising model.

Key words： magnetotellurics time series classification artificial neural network deep learning noise

Received: 10 April 2023 Published: 16 April 2024

ZTFLH:

P631

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Kai YANG
	Cheng LIU
	Jing-Long HE
	Han Li
	Chuan YAO

Cite this article:

Kai YANG,Cheng LIU,Jing-Long HE, et al. Research on magnetotelluric time series classification based on artificial neural network[J]. Geophysical and Geochemical Exploration, 2024, 48(2): 498-507.

URL:

https://www.wutanyuhuatan.com/EN/10.11720/wtyht.2024.1144 OR https://www.wutanyuhuatan.com/EN/Y2024/V48/I2/498

Basic structure of LSTM

LSTM structure diagram

FCN structure diagram

17]
">

Residual learning: A building block^[17]

Residual block in this paper

ResNet structure diagram

LSTM-FCN structure diagram

LSTM-ResNet structure diagram

The partial time series of training set

Training curve of each network

Comparison table of network parameters

The partial time series of measured data set

Comparison of classification accuracy of measured data sets

Schematic diagram of denoising system

Comparison of denoising results for simulated noisy data (partial)

[1]	孙洁, 晋光文, 白登海, 等. 大地电磁测深资料的噪声干扰[J]. 物探与化探, 2000, 24(2):119-127.
[1]	Sun J, Jin G W, Bai D H, et al. The noise interference of magnetotelluric sounding data[J]. Geophysical and Geochemical Exploration, 2000, 24(2):119-127.
[2]	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.
[2]	Yan J B. Research on theory and method of magnetotelluric signal processing[D]. Changsha: Central South University, 2003.
[3]	蔡剑华. 基于Hilbert-Huang变换的大地电磁信号处理方法与应用研究[D]. 长沙: 中南大学, 2010.
[3]	Cai J H. Research on magnetotelluric signal processing method and application based on Hilbert-Huang transform[D]. Changsha: Central South University, 2010.
[4]	周聪, 汤井田, 原源, 等. 强干扰区含噪电磁场的时空分布特征[J]. 吉林大学学报:地球科学版, 2020, 50(6):1870-1886.
[4]	Zhou C, Tang J T, Yuan Y, et al. Spatial and temporal distribution characteristics of electromagnetic fields in strong noise area[J]. Journal of Jilin University:Earth Science Edition, 2020, 50(6):1870-1886.
[5]	葛双超, 李斌. 大地电磁法人文噪声干扰特点及处理方法综述[J]. 物探化探计算技术, 2021, 43(5):609-619.
[5]	Ge S C, Li B. Review of the characteristics and processing methods of human noise interference in magnetotelluric[J]. Computing Techniques for Geophysical and Geochemical Exploration, 2021, 43(5):609-619.
[6]	杨一鸣, 潘嵘, 潘嘉林, 等. 时间序列分类问题的算法比较[J]. 计算机学报, 2007, 30(8):1259-1266.
[6]	Yang Y M, Pan R, Pan J L, et al. A comparative study on time series classification[J]. Chinese Journal of Computers, 2007, 30(8):1259-1266.
[7]	杨梦晨, 陈旭栋, 蔡鹏, 等. 早期时间序列分类方法研究综述[J]. 华东师范大学学报:自然科学版, 2021(5):115-133.
[7]	Yang M C, Chen X D, Cai P, et al. Survey of early time series classification methods[J]. Journal of East China Normal University:Natural Science, 2021(5):115-133.
[8]	沈旭东. 基于深度学习的时间序列算法综述[J]. 信息技术与信息化, 2019(1):71-76.
[8]	Shen X D. A survey of time series algorithms based on deep learning[J]. Information Technology and Informatization, 2019(1):71-76.
[9]	Mehak K. 时间序列分类的端到端深度学习方法[D]. 哈尔滨: 哈尔滨工业大学, 2021.
[9]	Mehak K. End-to-end deep learning method for time series classification[D]. Harbin:Harbin Institute of Technology, 2021.
[10]	Manoj C, Nagarajan N. The application of artificial neural networks to magnetotelluric time-series analysis[J]. Geophysical Journal International, 2003, 153(2):409-423.
[11]	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.
[11]	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.
[12]	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.
[12]	Wang S H, He L F. Feasibility analysis of the LSTM based denoising to MT time series[C]// The 9th Conference of Geophysical Technology,Chinese Geophysical Society—Abstracts of Symposium on Full-Space Geophysical Exploration and Intelligent Sensing, 2021.
[13]	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.
[13]	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.
[14]	Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780.
[15]	Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.
[16]	Wang Z G, Yan W Z, Oates T. Time series classification from scratch with deep neural networks:A strong baseline[C]// 2017 International Joint Conference on Neural Networks (IJCNN).Anchorage,AK,USA.IEEE, 2017:1578-1585.
[17]	He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas,NV,USA.IEEE, 2016:770-778.
[18]	刘晨, 赵晓晖, 梁乃川, 等. 基于ResNet50和迁移学习的岩性识别与分类研究[J]. 计算机与数字工程, 2021, 49(12):2526-2530,2578.
[18]	Liu C, Zhao X H, Liang N C, et al. Research on lithology identification and classification based on ResNet50 and transfer learning[J]. Computer & Digital Engineering, 2021, 49(12):2526-2530,2578.
[19]	张雅雯. 基于残差网络的时间序列的分类算法研究[D]. 北京: 北京交通大学, 2020.
[19]	Zhang Y W. Research on classification algorithm of time series based on residual network[D]. Beijing: Beijing Jiaotong University, 2020.
[20]	Karim F, Majumdar S, Darabi H, et al. LSTM fully convolutional networks for time series classification[J]. IEEE Access, 2017, 6:1662-1669.
[21]	Karim F, Majumdar S, Darabi H, et al. Multivariate LSTM-FCNs for time series classification[J]. Neural Networks, 2019, 116:237-245.
[22]	王琛, 王颖, 郑涛, 等. 基于ResNet-LSTM网络和注意力机制的综合能源系统多元负荷预测[J]. 电工技术学报, 2022, 37(7):1789-1799.
[22]	Wang C, Wang Y, Zheng T, et al. Multi-energy load forecasting in integrated energy system based on ResNet-LSTM network and attention mechanism[J]. Transactions of China Electrotechnical Society, 2022, 37(7):1789-1799.
[23]	叶兴扬, 韦张跃昊, 严卉, 等. 基于ResNet和Bi-LSTM模型融合的心电信号分类[J]. 航天医学与医学工程, 2021, 34(3):244-251.
[23]	Ye X Y, Wei Z Y H, Yan H, et al. ECG signal classification based on fusion model of res nte and Bi-LSTM network[J]. Space Medicine & Medical Engineering, 2021, 34(3):244-251.
[24]	Kingma D P, Ba J. Adam:A method for stochastic optimization[J]. ArXiv e-Prints,2014:arXiv:1412.6980.
[25]	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.
[25]	Yang K, Tang W D, Liu C, 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.