Seismic fault interpretation based on deep convolutional neural networks
CHANG Dekuan1,2, YONG Xueshan2, WANG Yihui2, YANG Wuyang2, LI Hai-shan2, ZHANG Guangzhi1
1. China University of Petroleum (East China), Qingdao, Shandong 266555, China; 2. Research Institute of Petroleum Exploration & Development-Northwest, PetroChina, Lanzhou, Gansu 730020, China
Abstract:Seismic fault interpretation has always been a key task in the process of oil and gas exploration and development. Conventional fault interpretation is mainly based on human-computer interaction, which is of low efficiency and causes the results with many uncertainties. In addition, conventional methods for fault interpretation usua-lly set multiple parameters, whose controls accuracy of the predicted faults. This paper proposes a method using seismic data based on convolutional deep neural networks. Taking the advantages of ResNet for effectively training deep convolutional neural network and U-Net architecture for characterizing multi-scale and multi-layer characteristic information, this method combines deep residual neural network and U-Net architecture to construct a network architecture (SeisFault-Net) for fault interpretation based on seismic data. The U-Net architecture consists of an encoding sub-network and a decoding sub-network. They enable the SeisFault-Net to train models in an end-to-end manner. The residual neural network can suppress the gradient dispersion of deep network, and effectively improve the training efficiency of the SeisFault-Net. After trained, the SeisFault-Net can perform fault interpretation based on seismic data without setting any parameters. This avoids the empirical error and uncertainties caused by parameters artificially set in conventional methods. Applications to raw data have proved that the SeisFault-Net me-thod can effectively and accurately detect fault loca-tions, and the faults have good vertical continuity and clear outlines. The detailed information of faults interpretated by the SeisFault-Net method is more abundant and accurate than the coherent algorithm. And the calculating efficiency of the SeisFault-Net method is very high in seismic fault interpretation.
Bahorich M,Farmer S.3-D seismic discontinuity for faults and stratigraphic features:The coherence cube[J].The Leading Edge,1995,14(10):1053-1058.
[2]
Marfurt K J,Kirlin R L,Farmer S L,et al. 3-D seismic attributes using a semblance-based coherency algorithm[J].Geophysics,1998,63(4):1150-1165.
[3]
Hale D.Methods to compute fault images,extract fault surfaces,and estimate fault throws from 3D seismic images[J].Geophysics,2013,78(2):O33-O43.
[4]
Marfurt K J,Sudhaker V,Gersztenkorn A,et al.Coherency calculations in the presence of structural dip[J].Geophysics,1999,64(1):104-111.
[5]
Pedersen S I,Randen T,Sonneland L,et al. Automatic fault extraction using artificial ants[C].SEG Technical Program Expanded Abstracts,2002,21:512-515.
[6]
Aurnhammer M,Tonnies K D.A genetic algorithm for automated horizon correlation across faults in seismic images[J].IEEE Transactions on Evolutionary Computation,2005,9(2):201-210.
[7]
印兴耀,高京华,宗兆云.基于离心窗倾角扫描的曲率属性提取[J].地球物理学报,2014,57(10):3411-3421.YIN Xingyao,GAO Jinghua,ZONG Zhaoyun. Curvature attribute based on dip scan with eccentric window[J].Chinese Journal of Geophysics,2014, 57(10):3411-3421.
[8]
陈雷,肖创柏,禹晶,等.基于相似性传播聚类与主成分分析的断层识别方法[J].石油地球物理勘探,2017,52(4):826-833.CHEN Lei,XIAO Chuangbai,YU Jing,et al. Fault recognition based on affinity propagation clustering and principal component analysis[J].Oil Geophysical Prospecting,2017,52(4):826-833.
[9]
赵凤全,崔德育,康婷婷,等.构造导向滤波技术在断裂识别中的应用[J].石油地球物理勘探,2018,53(增刊1):214-218.ZHAO Fengquan, CUI Deyu, KANG Tingting,et al. Fault identification with structure-oriented filtering[J]. Oil Geophysical Prospecting,2018,53(S1):214-218.
[10]
Wu X M,Hale D.3D seismic image processing for faults[J].Geophysics,2016,81(2):IM1-IM11.
[11]
Srisutthiyakorn N.Deep-learning methods for predicting permeability from 2D/3D binary-segmented images[C]. SEG Technical Program Expanded Abstracts,2016,35:3042-3046.
[12]
Lewis W,Vigh D.Deep learning prior models from seismic images for full-waveform inversion[C].SEG Technical Program Expanded Abstracts,2017,36:1512-1517.
[13]
Huang L,Dong X,Clee T E.A scalable deep learning platform for identifying geologic features from seismic attributes[J].The Leading Edge,2017,36(3):249-256.
[14]
Serfaty Y,Itan L,Chase D,et al.Wavefield separation via principle component analysis and deep learning in the local angle domain[C].SEG Technical Program Expanded Abstracts,2017,36:991-995.
[15]
Chang D K,Yang W Y,Yong X S,et al. Generative adversarial networks for seismic data interpolation[C]. SEG 2018 Workshop, SEG Maximizing Asset Value Through Artificial Intelligence and Machine Learning,Beijing,China,2018.
[16]
Chang D K,Yang W Y,Yong X S,et al. Seismic data interpolation using dual-domain conditional generative adversarial networks[J].IEEE Geoscience and Remote Sensing Letters,2020,99:1-5.
[17]
Wang B F,Zhang N,Lu W K,et al. Deep-learning-based seismic data interpolation:A preliminary result[J].Geophysics,2019,84(1):V11-V20.
[18]
王钰清,陆文凯,刘金林,等.基于数据增广和CNN的地震随机噪声压制[J].地球物理学报,2019,62(1):421-433.WANG Yuqing,LU Wenkai,LIU Jinlin,et al.Random seismic noise attenuation based on data augmentation and CNN[J].Chinese Journal of Geophysics,2019,62(1):421-433.
[19]
张玉玺,刘洋,张浩然,等.基于深度学习的多属性盐丘自动识别方法[J].石油地球物理勘探,2020,55(3):475-483.ZHANG Yuxi, LIU Yang, ZHANG Haoran,et al. Multi-attribute automatic interpretation of salt domes based on deep learning[J]. Oil Geophysical Prospecting, 2020, 55(3):475-483.
[20]
李海山,陈德武,吴杰,等.叠前随机噪声深度残差网络压制方法[J].石油地球物理勘探,2020,55(3):493-503.LI Haishan, CHEN Dewu, WU Jie, et al.Pre-stack random noise suppression with deep residual network[J].Oil Geophysical Prospecting,2020,55(3):493-503.
Wu X M,Shi Y Z,Fomel S,et al.Convolutional neural networks for fault interpretation in seismic images[C]. SEG Technical Program Expanded Abstracts,2018,37:1946-1950.
[23]
Chang D,Yong X,Yang W, et al.U_Net & residual neural networks for seismic fault interpretation[C].Extended Abstracts of 81st EAGE Conference and Exhibition,2019,1-5.
[24]
He K M,Zhang X,Ren S,et al.Identity mappings in deep residual networks[C].European Conference on Computer Vision,2016,630-645.
[25]
Nair V,Hinton G E.Rectified linear units improve restricted Boltzmann machines[C].International Conference on Machine Learning,2010,807-814.
[26]
Ronneberger O,Fischer P,Brox T. U-Net:convolutional networks for biomedical image segmentation[C]. Medical Image Computing and Computer-assisted Intervention-MICCAI 2015,Springer,2015,234-241.