Pub Date : 2019-12-09DOI: 10.3997/2214-4609.2019X610105
D. Egorov
Summary These days understanding of fault geometry distribution across a particular oil or gas reservoir becomes very important task. It arises from the fact that fluid flow of present unconventional deposits is mostly driven by natural fractures instead of sedimentary porosity and corresponding permeability. On the other, complex compartmentalized reservoir separated into small discontinuous deposits by tectonic activity could lead to economic risks during field development. Most of conventional tools for fault interpretation from seismic data are highly affected by noise from data and deterministic so cannot produce probabilistic output. In the presented research application of convolutional neural networks for fault interpretation from seismic data was considered. Proposed architecture and training process were described. It was shown by metrics and visual analysis that developed method is able to delineate faults from seismic data in different geological and geophysical conditions. Additional advantage of suggested approach is its ability to produce probabilistic output allowing robust work with geological uncertainties and economic risks related to them due to consideration of many probable cases.
{"title":"Automatic fault interpretation from seismic data via convolutional neural networks","authors":"D. Egorov","doi":"10.3997/2214-4609.2019X610105","DOIUrl":"https://doi.org/10.3997/2214-4609.2019X610105","url":null,"abstract":"Summary These days understanding of fault geometry distribution across a particular oil or gas reservoir becomes very important task. It arises from the fact that fluid flow of present unconventional deposits is mostly driven by natural fractures instead of sedimentary porosity and corresponding permeability. On the other, complex compartmentalized reservoir separated into small discontinuous deposits by tectonic activity could lead to economic risks during field development. Most of conventional tools for fault interpretation from seismic data are highly affected by noise from data and deterministic so cannot produce probabilistic output. In the presented research application of convolutional neural networks for fault interpretation from seismic data was considered. Proposed architecture and training process were described. It was shown by metrics and visual analysis that developed method is able to delineate faults from seismic data in different geological and geophysical conditions. Additional advantage of suggested approach is its ability to produce probabilistic output allowing robust work with geological uncertainties and economic risks related to them due to consideration of many probable cases.","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114477669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-09DOI: 10.3997/2214-4609.2019x610102
J. Downton, O. Collet, T. Colwell
Summary The challenge in adopting neural networks in the geosciences is the relative scarcity of labeled training data. This presentation demonstrates an approach to augment the amount of data used to train the neural network. Rock Physics theory is used to model the elastic parameter response due to changes in the rock and fluid properties of the local well control to generate a large number of pseudo wells. These pseudo wells are then used to model synthetic seismic gathers which are then used to train a Deep Neural Network (DNN). The trained DNN is then applied to the real dataset. Application of this workflow is shown for seismic reservoir characterization on a field in the North Sea producing commercial volumes of oil. The results are shown to have good continuity, are high in resolution which is compared to the prestack inversion approach.
{"title":"Rock-physics based Augmented Machine Learning for Reservoir Characterization","authors":"J. Downton, O. Collet, T. Colwell","doi":"10.3997/2214-4609.2019x610102","DOIUrl":"https://doi.org/10.3997/2214-4609.2019x610102","url":null,"abstract":"Summary The challenge in adopting neural networks in the geosciences is the relative scarcity of labeled training data. This presentation demonstrates an approach to augment the amount of data used to train the neural network. Rock Physics theory is used to model the elastic parameter response due to changes in the rock and fluid properties of the local well control to generate a large number of pseudo wells. These pseudo wells are then used to model synthetic seismic gathers which are then used to train a Deep Neural Network (DNN). The trained DNN is then applied to the real dataset. Application of this workflow is shown for seismic reservoir characterization on a field in the North Sea producing commercial volumes of oil. The results are shown to have good continuity, are high in resolution which is compared to the prestack inversion approach.","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133659163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-09DOI: 10.3997/2214-4609.2019X610101
A. J. Bugge, J. Lie
Summary The state-of-the-art seismic interpretation workflow is based on extraction of information from seismic images, which typically involves manual mapping of seed points along targeted geological structures. This process requires expert geophysical knowledge, interpretive experience, intuition and creativity. With increasing computational power, data science is continuously evolving and provide new digital tools applicable to various disciplines, including geoscience. Most of these tools are based on open-source signal processing, image processing and machine learning algorithms. By utilizing these digital tools and automate the extraction of information from seismic images, we can accumulate knowledge and build a subsurface understanding faster and better. Here, we introduce data-driven methods based on both supervised and unsupervised machine learning to address key aspects of an automated seismic interpretation workflow. We automatically identify and extract faults using a pre-trained conditional generative adaptive network together with image processing such as morphological operations. Further, we address stratigraphic units with a new 3D texture descriptor for seismic data, and compute and cluster feature vectors that describe seismic stratigraphy for given seismic sub-volumes. Finally, we correlate dislocated and truncated seismic horizons we introduce a non-local trace matching approach.
{"title":"Aspects of automated seismic interpretation using supervised and unsupervised machine learning","authors":"A. J. Bugge, J. Lie","doi":"10.3997/2214-4609.2019X610101","DOIUrl":"https://doi.org/10.3997/2214-4609.2019X610101","url":null,"abstract":"Summary The state-of-the-art seismic interpretation workflow is based on extraction of information from seismic images, which typically involves manual mapping of seed points along targeted geological structures. This process requires expert geophysical knowledge, interpretive experience, intuition and creativity. With increasing computational power, data science is continuously evolving and provide new digital tools applicable to various disciplines, including geoscience. Most of these tools are based on open-source signal processing, image processing and machine learning algorithms. By utilizing these digital tools and automate the extraction of information from seismic images, we can accumulate knowledge and build a subsurface understanding faster and better. Here, we introduce data-driven methods based on both supervised and unsupervised machine learning to address key aspects of an automated seismic interpretation workflow. We automatically identify and extract faults using a pre-trained conditional generative adaptive network together with image processing such as morphological operations. Further, we address stratigraphic units with a new 3D texture descriptor for seismic data, and compute and cluster feature vectors that describe seismic stratigraphy for given seismic sub-volumes. Finally, we correlate dislocated and truncated seismic horizons we introduce a non-local trace matching approach.","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"269 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132182670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-12-09DOI: 10.3997/2214-4609.2019X610110
B. Alaei, S. Purves, E. Larsen, D. Economou, D. Austin
Summary The history of hydrocarbon exploration consistently indicates the advantages of integrating knowledge and data derived from different disciplines such as basin modelling, structural geology and geophysics. We have designed a complete subsurface workflow or platform, that we call Digital Underground. It combines semi-automated data wrangling, highly accessible structured analytics ready data in large databases, direct integration of data analytics and machine learning technology, tracking of data provenance, enabling reproducible scientific workflows, and the practical use of ML methods by the geoscientist in making their decisions. The workflow uses ML approaches at different scales, from core to seismic, and basin to prospect scale; while providing dynamic access to large amounts of data throughout. The workflow includes four main stages starting with well data analysis and ending up in integration of data-driven distributions of different properties required for risk and volumetric estimations together with corresponding uncertainties. We have shown the advantage of the platform by testing it on several examples. ML technology paired with solid data science practice; facilitates the integration of data and disciplines, enables geoscientists to exceed current best practice with the ML tools, and paves the way to the "new" best practice which is integrated data science and geoscience.
{"title":"The Digital Underground: Integrating petroleum geoscience with data science principles to create an intelligent subsurface platform","authors":"B. Alaei, S. Purves, E. Larsen, D. Economou, D. Austin","doi":"10.3997/2214-4609.2019X610110","DOIUrl":"https://doi.org/10.3997/2214-4609.2019X610110","url":null,"abstract":"Summary The history of hydrocarbon exploration consistently indicates the advantages of integrating knowledge and data derived from different disciplines such as basin modelling, structural geology and geophysics. We have designed a complete subsurface workflow or platform, that we call Digital Underground. It combines semi-automated data wrangling, highly accessible structured analytics ready data in large databases, direct integration of data analytics and machine learning technology, tracking of data provenance, enabling reproducible scientific workflows, and the practical use of ML methods by the geoscientist in making their decisions. The workflow uses ML approaches at different scales, from core to seismic, and basin to prospect scale; while providing dynamic access to large amounts of data throughout. The workflow includes four main stages starting with well data analysis and ending up in integration of data-driven distributions of different properties required for risk and volumetric estimations together with corresponding uncertainties. We have shown the advantage of the platform by testing it on several examples. ML technology paired with solid data science practice; facilitates the integration of data and disciplines, enables geoscientists to exceed current best practice with the ML tools, and paves the way to the \"new\" best practice which is integrated data science and geoscience.","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126882596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.3997/2214-4609.2019x610115
S. A. Syed, T. Turkistani, M. Khan
{"title":"Data Driven Approach to Image, Classify and extract Seismic Discontinuities in Complex Geological Settings","authors":"S. A. Syed, T. Turkistani, M. Khan","doi":"10.3997/2214-4609.2019x610115","DOIUrl":"https://doi.org/10.3997/2214-4609.2019x610115","url":null,"abstract":"","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123689771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.3997/2214-4609.2019x610114
N. Bukhanov, A. Orlov, M. Ozhgibesov, E. Grishnyaev, T. Dogadova, B. Belozerov
{"title":"Data-driven well placement strategy based on variational simulations","authors":"N. Bukhanov, A. Orlov, M. Ozhgibesov, E. Grishnyaev, T. Dogadova, B. Belozerov","doi":"10.3997/2214-4609.2019x610114","DOIUrl":"https://doi.org/10.3997/2214-4609.2019x610114","url":null,"abstract":"","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134377170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.3997/2214-4609.2019x610116
A. A. Reshytko, D. Egorov, A. Klenitskiy, A. Shchepetnov
{"title":"WellNet: improvement of machine learning models robustness via comprehensive multi oilfield dataset","authors":"A. A. Reshytko, D. Egorov, A. Klenitskiy, A. Shchepetnov","doi":"10.3997/2214-4609.2019x610116","DOIUrl":"https://doi.org/10.3997/2214-4609.2019x610116","url":null,"abstract":"","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130374992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.3997/2214-4609.2019x610113
I. Fernandes, K. Mosegaard
{"title":"Quantification of errors in well-trace positions and uncertain measurements for improvement of subsurface imaging","authors":"I. Fernandes, K. Mosegaard","doi":"10.3997/2214-4609.2019x610113","DOIUrl":"https://doi.org/10.3997/2214-4609.2019x610113","url":null,"abstract":"","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133158229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.3997/2214-4609.2019x610104
G. Kocsis, H. Jaglan
{"title":"Pseudo-Wells based HitCube ‘trace-matching’ and Machine Learning Inversions: Seismic Reservoir Characterization in a Challenging Environment","authors":"G. Kocsis, H. Jaglan","doi":"10.3997/2214-4609.2019x610104","DOIUrl":"https://doi.org/10.3997/2214-4609.2019x610104","url":null,"abstract":"","PeriodicalId":369295,"journal":{"name":"EAGE Subsurface Intelligence Workshop","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129578434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：