Intelligent recognition method of low-grade faults based on VNet deep learning architecture
LU Pengfei1, DU Wenlong2, LI Li3, CHENG Danhua4, GUO Aihua1
1. School of Information Engineering, East China University of Technology, Nanchang, Jiangxi 330013, China;
2. The Third Exploration Team of Shandong Coal-field Geology Bureau, Tai'an, Shandong 271000, China;
3. Exploration and Development Research Institute of Qinghai Oilfield Company, PetroChina, Dunhuang, Gansu 736202, China;
4. Exploration and Development Research Institute of Jidong Oilfield Company, PetroChina, Tang-shan, Hebei 063004, China
Abstract:The recognition of low-grade faults is an important link in oil and gas exploration and development. Coherent volume, spectral decomposition, curvature, aunt body, edge detection, and other traditional methods have greatly improved the effect and accuracy of fault recognition, but they cannot effectively recognize low-grade faults with small fault distances. However, as an artificial intelligence technology, the deep learning method based on a full convolution neural network provides a new way for low-grade fault recognition. Based on UNet, the proposed VNet deep learning architecture can increase the receptive field of signals during the up and down sampling, extract large-scale fault information as much as possible, yet retain and extract small-scale fault information at the same time. Furthermore, this paper uses forward modeling data and actual seismic data to test UNet and VNet models, selects appropriate loss function, iteration times, and model weight parameters to compare the effects of model training and fault recognition. The results show that the VNet-based method can extract rich information and is more effective in low-grade fault recognition.
NEVES F A,ZAHRANI M S,BREMKAMP S W. Detection of potential fractures and small faults using seismic attributes[J].The Leading Edge,2004,23(9): 903-906.
[2]
罗群,黄捍东,王保华,等. 低序级断层的成因类型特征与地质意义[J]. 油气地质与采收率,2007,14(3): 19-21.LUO Qun,HUANG Handong,WANG Baohua,et al. Genetic types of low-grade faults and their geologic significance[J]. Petroleum Geology and Recovery Efficiency,2007,14(3): 19-21.
[3]
徐亮. 低序级断层的地震识别技术在惠民凹陷辛34断块中的应用[J]. 矿物岩石,2007,27(2): 86-93.XU Liang. Application of seismic discrimination technique to the low order fault in the 34 fault block of the Huiming sag[J]. Journal of Mineralogy and Petrology,2007,27(2): 86-93.
[4]
夏冰. 地震新技术在低级序断层识别中的应用[J]. 断块油气田,2007,14(2): 24-26.XIA Bing. Application of new seismic technique in identification of low-grade fault[J]. Fault-Block Oil & Gas Field,2007,14(2): 24-26.
[5]
边树涛,董艳蕾,苏晓军,等. 地震相干体技术识别低序级断层方法研究[J]. 世界地质,2007,26(3): 368-374.BIAN Shutao,DONG Yanlei,SU Xiaojun,et al. Method study of seismic coherence cube technique to interpretation of low level faults[J]. Global Geology,2007,26(3): 368-374.
[6]
边树涛,孟利,张柏枫.利用多尺度边缘检测技术识别低序级断层——以辛15断块古近系东营组为例[J]. 内蒙古石油化工,2011,37(12): 1-4.BIAN Shutao,MENG Li,ZHANG Baifeng. Using multi-scale edge detection technology to identify low sequence faults: a case study of Paleogene Dongying Formation in Xin 15 fault block[J].Inner Mongolia Petrochemical Industry,2011,37(12): 1-4.
[7]
CHOPRA S,MARFURT K J. Detecting stratigraphic features via crossplotting of seismic discontinuity attributes and their volume visualization[J].The Leading Edge,2009,28(12): 1422-1426.
[8]
李军,张军华,金强,等. 曲率属性在低序级断层识别中的应用[C]. 中国地球物理学会第二十八届年会论文集,2012,353.LI Jun,ZHANG Junhua,JIN Qiang,et al. The application of curvature attribute in identifying the low level faults[C]. Proceedings of the 28th Annual Meeting of the Chinese Geophysical Society,2012,353.
[9]
YU J B. Using cylindrical surface-based curvature change rate to detect faults and fractures[J]. Geophysics,2014,79(5): 1-9.
[10]
AHMAD M N,ROWELL P. Application of spectral decomposition and seismic attributes to understand the structure and distribution of sand reservoirs within Tertiary rift basins of the Gulf of Thailand[J].The Leading Edge,2012,31(6): 630-634.
[11]
JIBRIN B W,RESTON T J,WESTBROOK G K. Application of volumetric seismic discontinuity attri-bute for fault detection: case study using deep-water Niger delta 3D seismic data[J].The Leading Edge,2013,32(4): 424-428.
[12]
陈波,魏小东,任敦占,等. 基于谱分解技术的小断层识别[J].石油地球物理勘探,2010,45(6): 890-894.CHEN Bo,WEI Xiaodong,REN Dunzhan,et al. Small fault identification based on spectrum decomposition technique[J].Oil Geophysical Prospecting,2010,45(6): 890-894.
[13]
张昕,甘利灯,刘文岭,等. 密井网条件下井震联合低级序断层识别方法[J].石油地球物理勘探,2012,47(3): 462-468.ZHANG Xin,GAN Lideng,LIU Wenling,et al. Joint well-seismic interpretation of low-grade faults in dense well pattern block[J]. Oil Geophysical Prospecting,2012,47(3): 462-468.
[14]
宋劲. 地震属性在小断层及裂缝发育带预测中的应用[J]. 矿业安全与环保,2014,41(2): 32-34,40.SONG Jin. Application of seismic attributes in prediction of small fault and fracture-developed zone[J]. Mining Safety & Environmental Protection,2014,41(2): 32-34,40.
[15]
KAZMI H S,ALAM A,KHAN M M. Seismic volume attributes from time-frequency analysis[J]. The Leading Edge,2015,34(1): 62-70.
[16]
WU X M,HALE D. 3D seismic image processing for unconformities[J]. Geophysics,2015,80(2): IM35-IM44.
[17]
夏波,高荣锦,史海涛. 低序级断层识别技术在稠油难采区块研究中的应用[J].中国矿业,2016,25(增刊2): 324-327.XIA Bo,GAO Rongjin,SHI Haitao.Application of low level fault identification technologies in the hard-to-recover reserves of heavy oil[J].China Mining Magazine,2016,25(S2): 324-327.
[18]
黄诚,李鹏飞,王腾宇,等. 地震属性分析技术在小断层识别中的应用[J].工程地球物理学报,2016,13(1): 41-45.HUANG Cheng,LI Pengfei,WANG Tengyu,et al. The application of seismic attribute analysis technology to the identification of small faults[J]. Chinese Journal of Engineering Geophysics,2016,13(1): 41-45.
[19]
RONNEBERGER O,FISCHER P,BROX T. U-Net: convolutional networks for biomedical image segmentation[C]. Medical Image Computing and Computer-Assisted Intervention (MICCAI),2015,234-241.
[20]
HUANG L,DONG X S,CLEE T E. A scalable deep learning platform for identifying geologic features from seismic attributes[J].The Leading Edge,2017,36(3): 249-256.
[21]
GUO B W,LIU L,LUO Y.Automatic seismic fault detection with convolutional neural network[C].International Geophysical Conference,2018,1786-1789.
WANG Z R,LI B,LIU N H,et al. Distilling know-ledge from an ensemble of convolutional neural networks for seismic fault detection[J]. IEEE Geoscience and Remote Sensing Letters,2020,19: 1-5.
[26]
ARAYA-POLO M,DAHLKE T,FROGNER C,et al. Automated fault detection without seismic processing[J]. The Leading Edge,2017,36(3): 208-214.
[27]
CHANG D K,YANG W Y,YONG X S,et al. Seismic fault detection using deep learning technology[C]. International Geophysical Conference,2018,1770-1773.
[28]
WU X M,LIANG L M,SHI Y Z,et al. FaultSeg3D: using synthetic data sets to train an end-to-end convolutional neural network for 3D seismic fault segmentation[J].Geophysics,2019,84(3): IM35-IM45.
[29]
ZHOU R S,CAI Y F,YU F C,et al. Seismic fault detection with iterative deep learning[C].SEG Technical Program Expanded Abstracts,2019,38: 2503-2507.
[30]
张政,严哲,顾汉明. 基于残差网络与迁移学习的断层自动识别[J].石油地球物理勘探,2020,55(5): 950-956.ZHANG Zheng,YAN Zhe,GU Hanming. Automatic fault recognition with residual network and transfer learning[J].Oil Geophysical Prospecting,2020,55(5): 950-956.
[31]
吴吉忠,何书梅,杨倩倩,等.基于全卷积神经网络(FCN)的低序级断层识别方法研究[C]. SPG/SEG南京2020年国际地球物理会议论文集,2020,1010-1012.WU Jizhong,HE Shumei,YANG Qianqian,et al. Research on low order fault recognition method based on full convolution neural network (FCN)[C]. Procee- dings of SPG/SEG Nanjing 2020 International Geophysical Conference,2020,1010-1012.
[32]
常德宽,雍学善,王一惠,等. 基于深度卷积神经网络的地震数据断层识别方法[J].石油地球物理勘探,2021,56(1): 1-8.CHANG Dekuan,YONG Xueshan,WANG Yihui,et al. Seismic fault interpretation based on deep convolutional neural networks[J].Oil Geophysical Prospecting,2021,56(1): 1-8.
[33]
杨午阳,杨佳润,陈双全,等. 基于U-Net深度学习网络的地震数据断层检测[J].石油地球物理勘探,2021,56(4): 688-697.YANG Wuyang,YANG Jiarun,CHEN Shuangquan,et al. Seismic data fault detection based on U-Net deep learning network[J].Oil Geophysical Prospecting,2021,56(4): 688-697.
MILLETARI F,NAVAB N,AHMADI S A. V-Net: fully convolutional neural networks for volumetric medical image segmentation[C]. 2016 Fourth International Conference on 3D Vision (3DV),2016,565-571.