A Res-UNet network-based method for borehole-to-surface electrical resistivity inversion
ZHOU Nan1(), WANG Zhi1(), FANG Si-Nan2, ZHANG Yu-Zhe1
1. School of Electronic Information and Electrical Engineering, Yangtze University, Jingzhou 434023, China 2. College of Geophysics and Petroleum Resources, Yangtze University, Wuhan 430100, China
Traditional resistivity inversion methods tend to rely on the initial inversion model selected, get stuck in local minima, and be time-consuming. To address these issues, this study proposed a real-time resistivity inversion method based on the Res-UNet neural network. First, a significantly expanded forward response dataset was generated using the Gmsh software. Then, inversion experiments were carried out based on appropriate network parameters determined according to data characteristics. The experimental results indicate that the Res-UNet algorithm can fully dig the data characteristics and rapidly produce resistivity images that align with the electrical properties of strata. The experiments on the dataset for resistivity forward modeling yielded a mean squared error between the predicted values and the forward responses of 0.019 44, and those on the test set yielded a mean squared error of 0.075 8, suggesting improved imaging results compared to traditional inversion methods. Furthermore, the proposed method achieved encouraging results in the inversion calculations of simulation models, enabling rapid and accurate inversion of the location and morphologies of subsurface anomalies while exhibiting a strong noise resistance. This study provides a new method and philosophy for mapping the relationship between resistivity data and the actual geoelectric structures.
Cai J T, Ruan B Y, Zhao G Z, et al. Two-dimensional modeling of complex resistivity using finite element method[J]. Chinese Journal of Geophysics, 2007, 50(6):1869-1876.
Su Z, Hu W B. Meshless algorithm in two-dimensional electromagnetic forward calculation[J]. Geophysical and Geochemical Exploration, 2012, 36(6):1024-1028,1039.
Wang Z, Fang S N, Jiang K Y, et al. Research on 3D hole-to-surface resistivity forward modeling and anomaly based on unstructured meshes[J]. Progress in Geophysics, 2022, 37(4):1620-1630.
Ren Z Y, Tang J T. Finite element modeling of 3-D DC resistivity using locally refined unstructured meshes[J]. Chinese Journal of Geophysics, 2009, 52(10):2627-2634.
[5]
Tang J T, Wang F Y, Ren Z Y. 2.5D DC resistivity modeling by adaptive finite-element with unstructured triangulation[J]. Chinese Journal of Geophysics, 2010, 53(3):708-716.
Zhang Y Z, Meng L, Wang Z. Forward modeling of well-ground direct current resistivity method for undulating terrain based on Gmsh[J]. Geophysical and Geochemical Exploration, 2022, 46(1):182-190.
[7]
Constable S C, Parker R L, Constable C G. Occam’s inversion:A practical algorithm for generating smooth models from electromagnetic sounding data[J]. Geophysics, 1987, 52(3):289-300.
[8]
Smith J T, Booker J R. Rapid inversion of two- and three-dimensional magnetotelluric data[J]. Chinese Journal of Geophysical Research:Solid Earth, 1991, 96(B3):3905-3922.
[9]
Chen X B, Zhao G Z, Tang J, et al. The adaptive regularized inversion algorithm(ARIA) for magnetotelluric data[J]. Chinese Journal of Geophysics, 2005, 48(4):1005-1016.
Huang J G, Ruan B Y. An analytical comparison between 2d and 3d inversions in dc resistivity sounding[J]. Geophysical and Geochemical Exploration, 2004, 28(5):447-450.
Liu B, Li S C, Li S C, et al. 3D electrical resistivity inversion with least-squares method based on equality constraint and its computation efficiency optimization[J]. Chinese Journal of Geophysics, 2012, 55(1):260-268.
Wang Z, Pan H P, Luo Y H, et al. 3-D hole-to-surface resistivity inversion with nonlinear conjugate gradients method under the constraint of inequality[J]. Progress in Geophysics, 2016, 31(1):360-370.
Liu Z Y, Pang Y H, Zhang F K, et al. Deep learning resistivity inversion method based on multi-scale edge features[J]. Rock and Soil Mechanics, 2023, 44(11):3299-3306.
Xia W H, Zhu Z H, Han Y J, et al. Intelligent identification and segmentation method of wellbore fractures in resistivity logging imaging map[J]. Oil Geophysical Prospecting, 2023, 58(5):1042-1052.
[16]
Anomohanran O, Orhiunu M E. Assessment of groundwater occurrence in Olomoro,Nigeria using borehole logging and electrical resistivity methods[J]. Arabian Journal of Geosciences, 2018, 11(9):1-9.
[17]
Park G, Park S, Kim J H. Estimating the existence probability of cavities using integrated geophysics and a neural network approach[J]. Computers and Geosciences, 2010, 36(9):1161-1167.
Gao M L, Yu S B, Zheng J B, et al. Research of resistivity imaging using neural network based on immune genetic algorithm[J]. Chinese Journal of Geophysics, 2016, 59(11):4372-4382.
[20]
Phueakim K, Amatyakul P, Vachiratienchai C. An attempt to use convolutional neural network to recover layered-earth structure from electrical resistivity tomography survey[J]. Journal of Physics:Conference Series. IOP Publishing, 2023, 2653(1):012044.
[21]
He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]// Las Vegas:2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).NV,USA.IEEE,2016:770-778.
[22]
Ronneberger O, Fischer P, Brox T. U-net:Convolutional networks for biomedical image segmentation[M]// Lecture Notes in Computer science. Cham: Springer International Publishing,2015:234-241.
[23]
Srivastava N, Hinton G, Krizhevsky A, et al. Dropout:A simple way to prevent neural networks from overfitting.[J]. Journal of Machine Learning Research, 2014, 15(1):1929-1958.
[24]
He K M, Zhang X Y, Ren S Q, et al. Delving deep into rectifiers:Surpassing human-level performance on imagenet classification[C]// Santiago:2015 IEEE International Conference on Computer Vision(ICCV).Chile.IEEE,2015:1026-1034.
