Inversion of gravity full tensor gradient data based on U-Rnet network
QI Rui1, LI Houpu2, HU Jiaxin3, LUO Sha4
1. Department of Basic Courses, Naval University of Engineering, Wuhan, Hubei 430033, China; 2. College of Electrical Engi-neering, Naval University of Engineering, Wuhan, Hubei 430033 China; 3. United Imaging Surgical Technology Co., Ltd, Shanghai 200100, China; 4. School of Mathematics and Physics, China University of Geosciences, Wuhan, Hubei 430074, China
Abstract：Gravity inversion is one of the important means to obtain the spatial structure and physical properties of underground geological bodies through surface information， and each gravity gradient component represents different geological body information. Gravity inversion combined with gravity gradient components can better reflect the shape and distribution of underground abnormal bodies. In this paper， a neural network-based algorithm for gravity full tensor data inversion is proposed. The U-Rnet network is applied to three-dimensional gravity full tensor data inversion. In order to test the effectiveness of the algorithm， six representative models are used for simulation experiments， and inversion results with clear boundaries and sparsity are obtained. Firstly， by comparing the inversion results of L2 and Tversky loss functions， it is found that the inversion results corresponding to Tversky loss functions can clearly represent the boundary position of the model. Then， by comparing the inversion results of different gradient tensor combinations， the results of four tests show diffe-rent inversion accuracy on three directions （x， y， z）， and the test 4 shows the lowest fitting error. Finally， the proposed method is applied to the FTG data of Vinton Salt Dome in Texas， USA， and the inversion results are consistent with the real geological information.
祁锐, 李厚朴, 胡佳心, 罗莎. 基于U-Rnet的重力全张量梯度数据反演[J]. 石油地球物理勘探, 2024, 59(2): 331-342.
QI Rui, LI Houpu, HU Jiaxin, LUO Sha. Inversion of gravity full tensor gradient data based on U-Rnet network. Oil Geophysical Prospecting, 2024, 59(2): 331-342.
孟嘉春,蔡喜楣.卫星重力梯度测量及其应用前景探讨[J].地球物理学报,1991,34(3):369-376.MENG Jiachun,CAI Ximei.Approach on satellite gravity gradiometry and its vistas of applications[J].Acta Geophysics Sinica,1991,34(3):369-376.
[2]
周文月,马国庆,侯振隆,等.重力全张量数据联合欧拉反褶积法研究及应用[J].地球物理学报,2017,60(12):4855-4865.ZHOU Wenyue,MA Guoqing,HOU Zhenlong,et al.The study on the joint Euler deconvolution method of full tensor gravity data[J].Chinese Journal of Geophy-sics,2017,60(12):4855-4865.
[3]
ZHDANOV M S,ELLIS R,MUKHERJEE S.Three-dimensional regularized focusing inversion of gravity gradient tensor component data[J].Geophysics,2004,69(4):925-937.
[4]
WU L,KE X P,HSU H,et al.Joint gravity and gra-vity gradient inversion for subsurface object detection[J].IEEE Geoscience and Remote Sensing Letters,2013,10(4):865-869.
[5]
CAPRIOTTI J,LI Y G.Gravity and gravity gradient data:Understanding their information content through joint inversions[C].SEG Technical Program Expanded Abstracts,2014,33:1329-1333.
[6]
秦朋波,黄大年.重力和重力梯度数据联合聚焦反演方法[J].地球物理学报,2016,59(6):2203-2224.QIN Pengbo,HUANG Danian.Integrated gravity and gravity gradient data focusing inversion[J].Chinese Journal of Geophysics,2016,59(6):2203-2224.
[7]
侯振隆,王恩德,周文纳,等.重力梯度欧拉反褶积及其在文顿盐丘的应用[J].石油地球物理勘探,2019,54(2):472-479.HOU Zhenlong,WANG Ende,ZHOU Wenna,et al.Euler deconvolution of gravity gradiometry data and the application in Vinton Dome[J].Oil Geophysical Prospecting,2019,54(2):472-479.
[8]
张岩,李新月,王斌,等.基于深度学习的鲁棒地震数据去噪[J].石油地球物理勘探,2022,57(1):12-25.ZHANG Yan,LI Xinyue,WANG Bin,et al.Robust seismic data denoising based on deep learning[J].Oil Geophysical Prospecting,2022,57(1):12-25.
[9]
马国庆,王泽坤,李丽丽.基于自注意力机制深度学习的重磁数据网格化和滤波方法[J].石油地球物理勘探,2022,57(1):34-42.MA Guoqing,WANG Zekun,LI Lili.Gridding and filtering method of gravity and magnetic data based on self-attention deep learning[J].Oil Geophysical Prospecting,2022,57(1):34-42.
[10]
唐杰,孟涛,韩盛元,等.基于多分辨率U-Net网络的地震数据断层检测方法[J].石油地球物理勘探,2021,56(3):436-445.TANG Jie,MENG Tao,HAN Shengyuan,et al.A fault detection method of seismic data based on MultiResU-Net[J].Oil Geophysical Prospecting,2021,56(3):436-445.
[11]
YANG Q G,HU X Y,LIU S,et al.3-D gravity inversion based on deep convolution neural networks[J].IEEE Geoscience and Remote Sensing Letters,2022,19:1-5.
[12]
张志厚,廖晓龙,曹云勇,等.基于深度学习的重力异常与重力梯度异常联合反演[J].地球物理学报,2021,64(4):1435-1452.ZHANG Zhihou,LIAO Xiaolong,CAO Yunyong,et al.Joint gravity and gravity gradient inversion based on deep learning[J].Chinese Journal of Geophysics,2021,64(4):1435-1452.
[13]
CHAURASIA A,CULURCIELLO E.LinkNet:exploiting encoder representations for efficient semantic segmentation[C].2017 IEEE Visual Communications and Image Processing (VCIP),2017, 1-4.
[14]
ALTINI N,PRENCIPE B,BRUNETTI A.A tversky loss-based convolutional neural network for liver vessels segmentation[C].International Conference on Intelligent Computing,2020, 342-354.
[15]
LI X,CHOUTEAU M.Three-dimensional gravity modeling in all space[J].Surveys in Geophysics,1998,19(4):339-368.
[16]
马国庆,杜晓娟,李丽丽.解释位场全张量数据的张量局部波数法及其与常规局部波数法的比较[J].地球物理学报,2012,55(7):2450-2461.MA Guoqing,DU Xiaojuan,LI Lili.Comparison of the tensor local wavenumber method with the conventional local wavenumber method for interpretation of total tensor data of potential fields[J].Chinese Journal of Geophysics,2012,55(7):2450-2461.
[17]
高秀鹤,黄大年.基于共轭梯度算法的重力梯度数据三维聚焦反演研究[J].地球物理学报,2017,60(4):1571-1583.GAO Xiuhe,HUANG Danian.Research on 3D focu-sing inversion of gravity gradient tensor data based on a conjugate gradient algorithm[J].Chinese Journal of Geophysics,2017,60(4):1571-1583.
[18]
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),2016, 770-778.
[19]
RADFORD A,METZ L,CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks [EB/OL].(2015-11-19)[2024-01-26].https://www.science-open.com/document?vid=d0480825-f6f6-4176-9f25-28d199b6f86a.
ZHOU Y X,YANG G Z.Normalization in training U-Net for 2-D biomedical semantic segmentation[J].IEEE Robotics and Automation Letters,2019,4(2):1792-1799.
[22]
HE K M,ZHANG X Y,REN S Q,et al.Delving deep into rectifiers:surpassing Human-Level performance on ImageNet classification[C].2015 IEEE International Conference on Computer Vision (ICCV),2015,1026-1034.
[23]
GLOROT X,BENGIO Y.Understanding the difficulty of training deep feedforward neural networks[C].Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics,2010,249-256.
[24]
FARQUHARSON C G.Constructing piecewise-constant models in multidimensional minimum-structure inversions[J].Geophysics,2008,73(1):K1-K9.
[25]
GENG M X,HUANG D N,YANG Q J,et al.3D inversion of airborne gravity-gradiometry data using Cokriging[J].Geophysics,2014,79(4):G37-G47.
[26]
COKER M O,BHATTACHARYA J P,MARFURT K J.Fracture patterns within mudstones on the flanks of a salt dome:syneresis or slumping?[J].Gulf Coast Association of Geological Societies Transactions,2007,57:125-137.
[27]
DAVIS K,LI Y G.3D joint inversion of gradient and total-field magnetic data[C].SEG Technical Program Expanded Abstracts,2010,19:1784-1788.
[28]
OLIVEIRA V C,BARBOSA V C F.3-D radial gra-vity gradient inversion[J].Geophysical Journal International,2013,195(2):883-902.
[29]
ENNEN C.Mapping Gas-charged Fault Blocks Around the Vinton Salt Dome,Louisiana Using Gravity Gradiometry Data[D].University of Houston,Houston,USA,2012.
[30]
QIN P B,HUANG D N,YUAN Y,et al.Integrated gravity and gravity gradient 3D inversion using the non-linear conjugate gradient[J].Journal of Applied Geophysics,2016,126:52-73.
