Two-dimensional magnetotelluric inversion method based on deep learning
Objective The inversion of magnetotelluric sounding data to improve the accuracy of data interpretation has always been an essential topic in magnetotelluric sounding. Methods To address the problems of traditional magnetotelluric inversion methods, such as the dependence of the initial model and th...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Bulletin of Geological Science and Technology
2024-03-01
|
| Series: | 地质科技通报 |
| Subjects: | |
| Online Access: | https://dzkjqb.cug.edu.cn/en/article/doi/10.19509/j.cnki.dzkq.tb20220471 |
| _version_ | 1827277584103112704 |
|---|---|
| author | Fang WANG Jie XIONG Huixiao TIAN Siping LI Jiashuai KANG |
| author_facet | Fang WANG Jie XIONG Huixiao TIAN Siping LI Jiashuai KANG |
| author_sort | Fang WANG |
| collection | DOAJ |
| description | Objective The inversion of magnetotelluric sounding data to improve the accuracy of data interpretation has always been an essential topic in magnetotelluric sounding. Methods To address the problems of traditional magnetotelluric inversion methods, such as the dependence of the initial model and the ease of falling into a local optimum, this paper proposes a magnetotelluric inversion method based on deep learning.The method begins with the design of the GoogLeNetINV neural network. Then, various geoelectric models are constructed, and apparent resistivity data are extracted via forward modelling in the TM mode, constituting the training dataset. Additionally, the neural network is trained with the training dataset, and the network parameters are adjusted. Finally, the apparent resistivity data are input into the trained GoogLeNetINV neural network to directly obtain the inversion result. Results The experimental results reveal that the location and resistivity data of the "unlearned" geoelectric model can be inverted quickly and accurately, and the model has good generalization ability. The use of noise data can still yield good inversion results and a certain anti-noise ability. Conclusion The neural network is applied to the field data processing of the Bendigo Zone, and the resistivity model derived through inversion is consistent with the seismic interpretation. Consequently, the magnetotelluric inversion method based on deep learning can effectively solve the magnetotelluric inversion problem. |
| first_indexed | 2024-04-24T07:34:36Z |
| format | Article |
| id | doaj.art-ee014248486d47e982fab3c793756d20 |
| institution | Directory Open Access Journal |
| issn | 2096-8523 |
| language | zho |
| last_indexed | 2024-04-24T07:34:36Z |
| publishDate | 2024-03-01 |
| publisher | Editorial Department of Bulletin of Geological Science and Technology |
| record_format | Article |
| series | 地质科技通报 |
| spelling | doaj.art-ee014248486d47e982fab3c793756d202024-04-20T10:18:33ZzhoEditorial Department of Bulletin of Geological Science and Technology地质科技通报2096-85232024-03-0143234435410.19509/j.cnki.dzkq.tb20220471dzkjtb-43-2-344Two-dimensional magnetotelluric inversion method based on deep learningFang WANG0Jie XIONG1Huixiao TIAN2Siping LI3Jiashuai KANG4School of Electronic Information, Yangtze University, Jingzhou Hubei 434023, ChinaSchool of Electronic Information, Yangtze University, Jingzhou Hubei 434023, ChinaSchool of Electronic Information, Yangtze University, Jingzhou Hubei 434023, ChinaSchool of Electronic Information, Yangtze University, Jingzhou Hubei 434023, ChinaSchool of Electronic Information, Yangtze University, Jingzhou Hubei 434023, ChinaObjective The inversion of magnetotelluric sounding data to improve the accuracy of data interpretation has always been an essential topic in magnetotelluric sounding. Methods To address the problems of traditional magnetotelluric inversion methods, such as the dependence of the initial model and the ease of falling into a local optimum, this paper proposes a magnetotelluric inversion method based on deep learning.The method begins with the design of the GoogLeNetINV neural network. Then, various geoelectric models are constructed, and apparent resistivity data are extracted via forward modelling in the TM mode, constituting the training dataset. Additionally, the neural network is trained with the training dataset, and the network parameters are adjusted. Finally, the apparent resistivity data are input into the trained GoogLeNetINV neural network to directly obtain the inversion result. Results The experimental results reveal that the location and resistivity data of the "unlearned" geoelectric model can be inverted quickly and accurately, and the model has good generalization ability. The use of noise data can still yield good inversion results and a certain anti-noise ability. Conclusion The neural network is applied to the field data processing of the Bendigo Zone, and the resistivity model derived through inversion is consistent with the seismic interpretation. Consequently, the magnetotelluric inversion method based on deep learning can effectively solve the magnetotelluric inversion problem.https://dzkjqb.cug.edu.cn/en/article/doi/10.19509/j.cnki.dzkq.tb20220471deep learningmagnetotelluric inversionneural networkgooglenetinv |
| spellingShingle | Fang WANG Jie XIONG Huixiao TIAN Siping LI Jiashuai KANG Two-dimensional magnetotelluric inversion method based on deep learning 地质科技通报 deep learning magnetotelluric inversion neural network googlenetinv |
| title | Two-dimensional magnetotelluric inversion method based on deep learning |
| title_full | Two-dimensional magnetotelluric inversion method based on deep learning |
| title_fullStr | Two-dimensional magnetotelluric inversion method based on deep learning |
| title_full_unstemmed | Two-dimensional magnetotelluric inversion method based on deep learning |
| title_short | Two-dimensional magnetotelluric inversion method based on deep learning |
| title_sort | two dimensional magnetotelluric inversion method based on deep learning |
| topic | deep learning magnetotelluric inversion neural network googlenetinv |
| url | https://dzkjqb.cug.edu.cn/en/article/doi/10.19509/j.cnki.dzkq.tb20220471 |
| work_keys_str_mv | AT fangwang twodimensionalmagnetotelluricinversionmethodbasedondeeplearning AT jiexiong twodimensionalmagnetotelluricinversionmethodbasedondeeplearning AT huixiaotian twodimensionalmagnetotelluricinversionmethodbasedondeeplearning AT sipingli twodimensionalmagnetotelluricinversionmethodbasedondeeplearning AT jiashuaikang twodimensionalmagnetotelluricinversionmethodbasedondeeplearning |