To solve the problems in restoring sedimentary facies and predicting reservoirs in loose gas-bearing sediment,based on seismic sedimentologic analysis of the first 9-component S-wave 3D seismic dataset of China,a four...To solve the problems in restoring sedimentary facies and predicting reservoirs in loose gas-bearing sediment,based on seismic sedimentologic analysis of the first 9-component S-wave 3D seismic dataset of China,a fourth-order isochronous stratigraphic framework was set up and then sedimentary facies and reservoirs in the Pleistocene Qigequan Formation in Taidong area of Qaidam Basin were studied by seismic geomorphology and seismic lithology.The study method and thought are as following.Firstly,techniques of phase rotation,frequency decomposition and fusion,and stratal slicing were applied to the 9-component S-wave seismic data to restore sedimentary facies of major marker beds based on sedimentary models reflected by satellite images.Then,techniques of seismic attribute extraction,principal component analysis,and random fitting were applied to calculate the reservoir thickness and physical parameters of a key sandbody,and the results are satisfactory and confirmed by blind testing wells.Study results reveal that the dominant sedimentary facies in the Qigequan Formation within the study area are delta front and shallow lake.The RGB fused slices indicate that there are two cycles with three sets of underwater distributary channel systems in one period.Among them,sandstones in the distributary channels of middle-low Qigequan Formation are thick and broad with superior physical properties,which are favorable reservoirs.The reservoir permeability is also affected by diagenesis.Distributary channel sandstone reservoirs extend further to the west of Sebei-1 gas field,which provides a basis to expand exploration to the western peripheral area.展开更多
Seismic inversion is one of the most important methods for lithological prospecting . Seismic data with lowresolution is converted into impedance data of high resolution which can reflect the geological structure by i...Seismic inversion is one of the most important methods for lithological prospecting . Seismic data with lowresolution is converted into impedance data of high resolution which can reflect the geological structure by inversionThe inversion technique of 3D seismic data is discussed from both methodological and theoretical aspects, and the in-version test is also carried out using actual logging data. The result is identical with the measured data obtained fromroadway of coal mine. The field tests and research results indicate that this method can provide more accurate data foridentifying thin coal seam and minor faults.展开更多
3D visualization technology is a tool used for displaying, describing, and understanding the characteristics of geologic bodies, and features high efficiency, objective accuracy, visual expression, etc. In this paper,...3D visualization technology is a tool used for displaying, describing, and understanding the characteristics of geologic bodies, and features high efficiency, objective accuracy, visual expression, etc. In this paper, the man-machine interactive interpretation and 3D visualization technology rapidly displaying and analyzing the 3D seismic data of hydrate ore volume is researched and developed using the hybrid rendering technique. Through the integrated interpretation on the 3D space structure, stratum, and seismic attributes, the visualized multi-attribute superimposition analysis is implemented for describing the spatial distribution characteristics of hydrate ore volume and exquisitely describing the subtle geological characteristics of hydrate ore volume. By the hybrid rendering technique, authentication and interpretation are provided for the geological exploration work, so as to greatly enhance the visualization and accuracy of the geological analysis, and also provide a good decision-making foundation for the subsequent development of resources.展开更多
This paper presents a new approach for attenuating coherent noise in 3D seismic data. An adaptive beamforming with generalized sidelobe canceller (GSC) design methodology is utilized here as a general form of linearly...This paper presents a new approach for attenuating coherent noise in 3D seismic data. An adaptive beamforming with generalized sidelobe canceller (GSC) design methodology is utilized here as a general form of linearly constrained adaptive beamforming structure. It consists of a fixed beamformer, and a signal-blocking matrix in front of an unconstrained adaptive beamformer.Considerationf of the complexity of the geometry for 3D seismic survey, the 3D beamforming with GSC technique is developed with two key points: (1) sorting along azimuth sections to simplify the relationship between traveltime and offset from 3D to 2D, and (2) dynamic binning scheme to avoid the possible poor folding in some azimuth sections. Both simulation result and real data example show that the newly developed 3D beamforming with GSC yields more credible results at a relative low cost, sufficient stability and good resolution.展开更多
The laser scanning and CCD image-transmitting measurement method and principle on acquiring 3-D curved surface shape data are discussed. Computer processing technique of 3-D curved surface shape(be called“ 3 - D surf...The laser scanning and CCD image-transmitting measurement method and principle on acquiring 3-D curved surface shape data are discussed. Computer processing technique of 3-D curved surface shape(be called“ 3 - D surface shape”for short) data is analysed. This technique in- cludes these concrete methods and principles such as data smoothing, fitting, reconstructing ,elimi- nating and so on. The example and result about computer processing of 3- D surface shape data are given .展开更多
The Lower Cambrian Longwangmiao Fm gas reservoir in the GaoshitieMoxi area,the Sichuan Basin,is a super giant monoblock marine carbonate gas reservoir with its single size being the largest in China.The key to the rea...The Lower Cambrian Longwangmiao Fm gas reservoir in the GaoshitieMoxi area,the Sichuan Basin,is a super giant monoblock marine carbonate gas reservoir with its single size being the largest in China.The key to the realization of high and stable production gas wells in this gas reservoir is to identify accurately high-permeability zones where there are dissolved pores or dissolved pores are superimposed with fractures.However,high quality dolomite reservoirs are characterized by large burial depth and strong heterogeneity,so reservoir prediction is of difficult.In this paper,related seismic researches were carried out and supporting technologies were developed as follows.First,a geologic model was built after an analysis of the existing data and forward modeling was carried out to establish a reservoir seismic response model.Second,by virtue of well-oriented amplitude processing technology,spherical diffusion compensation factor was obtained based on VSP well logging data and the true amplitude of seismic data was recovered.Third,the resolution of deep seismic data was improved by using the welloriented high-resolution frequency-expanding technology and prestack time migration data of high quality was acquired.And fourth,multiple shoal facies reservoirs were traced by using the global automatic seismic interpretation technology which is based on stratigraphic model,multiple reservoirs which are laterally continuous and vertically superimposed could be predicted,and the areal distribution of high quality reservoirs could be described accurately and efficiently.By virtue of the supporting technologies,drilling trajectory is positioned accurately,and the deployed development wells all have high yield.These technologies also promote the construction of a modern supergiant gas field of tens of billions of cubic meters.展开更多
The northern Sichuan Basin,spreading in the northern section of the Longmenshan Fault Fold Belt,is characterized by dramatic fluctuations in the surface landforms,development of abdominal faults,and low-quality seismi...The northern Sichuan Basin,spreading in the northern section of the Longmenshan Fault Fold Belt,is characterized by dramatic fluctuations in the surface landforms,development of abdominal faults,and low-quality seismic data,resulting in difficulties in clarifying relevant structures.The key target formation,the Mid-Permian Qixia Fm,is deeply buried with thin reservoirs and high heterogeneity,which brings great challenges to seismic prediction.Under such circumstances,researches have been conducted jointly in terms of seismic data acquisition,processing and interpretation,and finally some relevant seismic survey technologies were developed suitable for surface/underground complex structures.Through surface structural surveys,dynamic deep-well lithologic identification,single-point detector deployment and process optimization,acquisition parameters can be excited.In addition,by using an observation systemwith high-coverage,wide-azimuth and huge-displacement,quality of acquired seismic data can be enhanced dramatically.Seismic imaging technologies for complex structures have been developed to enhance the quality of images for deep formations.These technologies are dominated by microscopic logging-constrained tomography static correction,high-resolution processing with fidelity and amplitude preservation and all-around PSDM in an angular domain.By using high-resolution gravity,magnetic and electric data,details related to geological structures and faults can be identified.In combination with fine seismic data interpretation,structural details and fault features can be verified effectively.Based on forwardmodeling and fine seismic calibration of reservoirs in individualwells,suitable attributes can be identified for predictions related to the distribution of reservoirs.By using all these auxiliary technologies,a large-scale structuralelithologic composite trap with a total area of 1223 km^(2) has been discovered in the NW Sichuan Basin.The Shuangyushi-Jiangyou area as a whole distributes on structural highs.In the areas to the south of Shuangyushi,the Qixia Fm dolomite reservoirs of platform margins are continuously developed.In conclusion,these auxiliary technologies can effectively allow trap identification and thin reservoir prediction in complex structures in the study area.In addition to clarifying the exploration orientation and providing a necessarily technical supports forwell development,these technologies help to accelerate the construction of demonstration projects for the exploration and development of deep marine carbonate formations.展开更多
3D seismic prospecting in mining areas of Xieqiao Colliery is a successfulmodel for an advancement from the resource prospecting to mining prospecting stagein coal fields. Its results have proved that faults with a th...3D seismic prospecting in mining areas of Xieqiao Colliery is a successfulmodel for an advancement from the resource prospecting to mining prospecting stagein coal fields. Its results have proved that faults with a throw of 5-10 m can be detected in an area with good seismogeologic conditions by using 3D seismic technique.Detection of underground tunnels for the first time utilizing 3D seismic data indicates that subsided columns, gotten and mine goaf can be detected using 3D seismic technique, so it has a broad applied prospect.展开更多
A typical building project has a long life in the maintenance stage. Also, the cost at this stage is enormously huge compared to planning, design and construction phases. In the earlier stage, which is planning or des...A typical building project has a long life in the maintenance stage. Also, the cost at this stage is enormously huge compared to planning, design and construction phases. In the earlier stage, which is planning or design phase, however, many project participants put little emphasis on the maintenance information. As a result, important maintenance data is missing and erroneously feedback to the 3D/BIM model. This research provides a generic process model for maintenance information management for building facilities. The authors have identified that there exist most-frequently used information areas: checking information, material information, equipment information, supplier information, and maintenance history information. Each information area should be embedded in the BIM model in order to effectively feedback to the operation and maintenance stage in the project. Thus, the study has proposed a novel data format structure which can effectively link the 3D/BIM object with the maintenance data. The demonstration project shows how the data format structure is used. The contribution of this study is to provide guidance to a project practitioner by step-by-step approach in dealing with the significant maintenance information in the earlier stage of the construction project.展开更多
According to the seismic and geological differences among every oil measures in mid-deep layers at west slope in Qikou Sag, varieties of new techniques on geophysics and geochemistry were introduced , such as seismic ...According to the seismic and geological differences among every oil measures in mid-deep layers at west slope in Qikou Sag, varieties of new techniques on geophysics and geochemistry were introduced , such as seismic pro-cess of target and CWS, neural network seismic microfacies cluster analysis, 3D interval velocity analysis. Taking advan-tage of the series of new techniques aims at predicting synthetically and quantitatively the deep gravity flow channel sandbody. Furthermore, considering the structural setting data, potential structural-lithologic traps were specified. As a re-sult, the geological and drilling effect is obviously promoted.展开更多
Turbidity currents represent a major agent for sediment transport in lakes, seas and oceans. In particu-lar, they formulate the most significant clastic accumulations in the deep sea, which become many of the world＇s...Turbidity currents represent a major agent for sediment transport in lakes, seas and oceans. In particu-lar, they formulate the most significant clastic accumulations in the deep sea, which become many of the world＇s most important hydrocarbon reservoirs. Several boreholes in the Qiongdongnan Basin, the north-western South China Sea, have recently revealed turbidity current deposits as significant hydrocarbon res-ervoirs. However, there are some arguments for the potential provenances. To solve this problem, it is es-sential to delineate their sedimentary processes as well as to evaluate their qualities as reservoir. Numerical simulations have been developed rapidly over the last several years, offering insights into turbidity current behaviors, as geologically significant turbidity currents are difficult to directly investigate due to their large scale and often destructive nature. Combined with the interpretation of the turbidity system based on high-resolution 3D seismic data, the paleotophography is acquired via a back-stripping seismic profile integrated with a borehole, i.e., Well A, in the western Qiongdongnan Basin; then a numerical model is built on the basis of this back-stripped profile. After defining the various turbidity current initial boundary conditions, includ-ing grain size, velocity and sediment concentration, the structures and behaviors of turbidity currents are investigated via numerical simulation software ANSYS FLUENT. Finally, the simulated turbidity deposits are compared with the interpreted sedimentary bodies based on 3D seismic data and the potential provenances of the revealed turbidites by Well A are discussed in details. The simulation results indicate that a sedimen-tary body develops far away from its source with an average grain size of 0.1 mm, i.e., sand-size sediment. Taking into account the location and orientation of the simulated seismic line, the consistence between normal forward simulation results and the revealed cores in Well A indicates that the turbidites should have been transported from Vietnam instead of Hainan Island. This interpretation has also been verified by the planar maps of sedimentary systems based on integration of boreholes and seismic data. The identification of the turbidity provenance will benefit the evaluation of extensively distributed submarine fans for hydro-carbon exploration in the deepwater areas.展开更多
Deep learning technologies are increasingly used in the fi eld of geophysics,and a variety of algorithms based on shallow convolutional neural networks are more widely used in fault recognition,but these methods are u...Deep learning technologies are increasingly used in the fi eld of geophysics,and a variety of algorithms based on shallow convolutional neural networks are more widely used in fault recognition,but these methods are usually not able to accurately identify complex faults.In this study,using the advantage of deep residual networks to capture strong learning features,we introduce residual blocks to replace all convolutional layers of the three-dimensional(3D)UNet to build a new 3D Res-UNet and select appropriate parameters through experiments to train a large amount of synthesized seismic data.After the training is completed,we introduce the mechanism of knowledge distillation.First,we treat the 3D Res-UNet as a teacher network and then train the 3D Res-UNet as a student network;in this process,the teacher network is in evaluation mode.Finally,we calculate the mixed loss function by combining the teacher model and student network to learn more fault information,improve the performance of the network,and optimize the fault recognition eff ect.The quantitative evaluation result of the synthetic model test proves that the 3D Res-UNet can considerably improve the accuracy of fault recognition from 0.956 to 0.993 after knowledge distillation,and the eff ectiveness and feasibility of our method can be verifi ed based on the application of actual seismic data.展开更多
基金Supported by the CNPC Science and Technology Projects(2022-N/G-47808,2023-N/G-67014)RIPED International Cooperation Project(19HTY5000008).
文摘To solve the problems in restoring sedimentary facies and predicting reservoirs in loose gas-bearing sediment,based on seismic sedimentologic analysis of the first 9-component S-wave 3D seismic dataset of China,a fourth-order isochronous stratigraphic framework was set up and then sedimentary facies and reservoirs in the Pleistocene Qigequan Formation in Taidong area of Qaidam Basin were studied by seismic geomorphology and seismic lithology.The study method and thought are as following.Firstly,techniques of phase rotation,frequency decomposition and fusion,and stratal slicing were applied to the 9-component S-wave seismic data to restore sedimentary facies of major marker beds based on sedimentary models reflected by satellite images.Then,techniques of seismic attribute extraction,principal component analysis,and random fitting were applied to calculate the reservoir thickness and physical parameters of a key sandbody,and the results are satisfactory and confirmed by blind testing wells.Study results reveal that the dominant sedimentary facies in the Qigequan Formation within the study area are delta front and shallow lake.The RGB fused slices indicate that there are two cycles with three sets of underwater distributary channel systems in one period.Among them,sandstones in the distributary channels of middle-low Qigequan Formation are thick and broad with superior physical properties,which are favorable reservoirs.The reservoir permeability is also affected by diagenesis.Distributary channel sandstone reservoirs extend further to the west of Sebei-1 gas field,which provides a basis to expand exploration to the western peripheral area.
文摘Seismic inversion is one of the most important methods for lithological prospecting . Seismic data with lowresolution is converted into impedance data of high resolution which can reflect the geological structure by inversionThe inversion technique of 3D seismic data is discussed from both methodological and theoretical aspects, and the in-version test is also carried out using actual logging data. The result is identical with the measured data obtained fromroadway of coal mine. The field tests and research results indicate that this method can provide more accurate data foridentifying thin coal seam and minor faults.
文摘3D visualization technology is a tool used for displaying, describing, and understanding the characteristics of geologic bodies, and features high efficiency, objective accuracy, visual expression, etc. In this paper, the man-machine interactive interpretation and 3D visualization technology rapidly displaying and analyzing the 3D seismic data of hydrate ore volume is researched and developed using the hybrid rendering technique. Through the integrated interpretation on the 3D space structure, stratum, and seismic attributes, the visualized multi-attribute superimposition analysis is implemented for describing the spatial distribution characteristics of hydrate ore volume and exquisitely describing the subtle geological characteristics of hydrate ore volume. By the hybrid rendering technique, authentication and interpretation are provided for the geological exploration work, so as to greatly enhance the visualization and accuracy of the geological analysis, and also provide a good decision-making foundation for the subsequent development of resources.
基金This research is sponsored by by China Natural Science Foundation (40274041), China National Petroleum Corporation (CNPC)Innovation Fund (2002CXKF-3)
文摘This paper presents a new approach for attenuating coherent noise in 3D seismic data. An adaptive beamforming with generalized sidelobe canceller (GSC) design methodology is utilized here as a general form of linearly constrained adaptive beamforming structure. It consists of a fixed beamformer, and a signal-blocking matrix in front of an unconstrained adaptive beamformer.Considerationf of the complexity of the geometry for 3D seismic survey, the 3D beamforming with GSC technique is developed with two key points: (1) sorting along azimuth sections to simplify the relationship between traveltime and offset from 3D to 2D, and (2) dynamic binning scheme to avoid the possible poor folding in some azimuth sections. Both simulation result and real data example show that the newly developed 3D beamforming with GSC yields more credible results at a relative low cost, sufficient stability and good resolution.
文摘The laser scanning and CCD image-transmitting measurement method and principle on acquiring 3-D curved surface shape data are discussed. Computer processing technique of 3-D curved surface shape(be called“ 3 - D surface shape”for short) data is analysed. This technique in- cludes these concrete methods and principles such as data smoothing, fitting, reconstructing ,elimi- nating and so on. The example and result about computer processing of 3- D surface shape data are given .
文摘The Lower Cambrian Longwangmiao Fm gas reservoir in the GaoshitieMoxi area,the Sichuan Basin,is a super giant monoblock marine carbonate gas reservoir with its single size being the largest in China.The key to the realization of high and stable production gas wells in this gas reservoir is to identify accurately high-permeability zones where there are dissolved pores or dissolved pores are superimposed with fractures.However,high quality dolomite reservoirs are characterized by large burial depth and strong heterogeneity,so reservoir prediction is of difficult.In this paper,related seismic researches were carried out and supporting technologies were developed as follows.First,a geologic model was built after an analysis of the existing data and forward modeling was carried out to establish a reservoir seismic response model.Second,by virtue of well-oriented amplitude processing technology,spherical diffusion compensation factor was obtained based on VSP well logging data and the true amplitude of seismic data was recovered.Third,the resolution of deep seismic data was improved by using the welloriented high-resolution frequency-expanding technology and prestack time migration data of high quality was acquired.And fourth,multiple shoal facies reservoirs were traced by using the global automatic seismic interpretation technology which is based on stratigraphic model,multiple reservoirs which are laterally continuous and vertically superimposed could be predicted,and the areal distribution of high quality reservoirs could be described accurately and efficiently.By virtue of the supporting technologies,drilling trajectory is positioned accurately,and the deployed development wells all have high yield.These technologies also promote the construction of a modern supergiant gas field of tens of billions of cubic meters.
基金Project supported by the National Major Science and Technology Project“Development of Large Oil/Gas Fields and Coalbed Methane”(No.2016ZX05004-005&2016ZX05007-004)the CNPC Major Science and Technology Project“Research and application of key technologies for maintaining gas productivity of 30 billion cubic meters in Southwest Oil and Gas Fields”(No.2016E-06,2016E-0602,2016E-0603&2016E-0604).
文摘The northern Sichuan Basin,spreading in the northern section of the Longmenshan Fault Fold Belt,is characterized by dramatic fluctuations in the surface landforms,development of abdominal faults,and low-quality seismic data,resulting in difficulties in clarifying relevant structures.The key target formation,the Mid-Permian Qixia Fm,is deeply buried with thin reservoirs and high heterogeneity,which brings great challenges to seismic prediction.Under such circumstances,researches have been conducted jointly in terms of seismic data acquisition,processing and interpretation,and finally some relevant seismic survey technologies were developed suitable for surface/underground complex structures.Through surface structural surveys,dynamic deep-well lithologic identification,single-point detector deployment and process optimization,acquisition parameters can be excited.In addition,by using an observation systemwith high-coverage,wide-azimuth and huge-displacement,quality of acquired seismic data can be enhanced dramatically.Seismic imaging technologies for complex structures have been developed to enhance the quality of images for deep formations.These technologies are dominated by microscopic logging-constrained tomography static correction,high-resolution processing with fidelity and amplitude preservation and all-around PSDM in an angular domain.By using high-resolution gravity,magnetic and electric data,details related to geological structures and faults can be identified.In combination with fine seismic data interpretation,structural details and fault features can be verified effectively.Based on forwardmodeling and fine seismic calibration of reservoirs in individualwells,suitable attributes can be identified for predictions related to the distribution of reservoirs.By using all these auxiliary technologies,a large-scale structuralelithologic composite trap with a total area of 1223 km^(2) has been discovered in the NW Sichuan Basin.The Shuangyushi-Jiangyou area as a whole distributes on structural highs.In the areas to the south of Shuangyushi,the Qixia Fm dolomite reservoirs of platform margins are continuously developed.In conclusion,these auxiliary technologies can effectively allow trap identification and thin reservoir prediction in complex structures in the study area.In addition to clarifying the exploration orientation and providing a necessarily technical supports forwell development,these technologies help to accelerate the construction of demonstration projects for the exploration and development of deep marine carbonate formations.
文摘3D seismic prospecting in mining areas of Xieqiao Colliery is a successfulmodel for an advancement from the resource prospecting to mining prospecting stagein coal fields. Its results have proved that faults with a throw of 5-10 m can be detected in an area with good seismogeologic conditions by using 3D seismic technique.Detection of underground tunnels for the first time utilizing 3D seismic data indicates that subsided columns, gotten and mine goaf can be detected using 3D seismic technique, so it has a broad applied prospect.
文摘A typical building project has a long life in the maintenance stage. Also, the cost at this stage is enormously huge compared to planning, design and construction phases. In the earlier stage, which is planning or design phase, however, many project participants put little emphasis on the maintenance information. As a result, important maintenance data is missing and erroneously feedback to the 3D/BIM model. This research provides a generic process model for maintenance information management for building facilities. The authors have identified that there exist most-frequently used information areas: checking information, material information, equipment information, supplier information, and maintenance history information. Each information area should be embedded in the BIM model in order to effectively feedback to the operation and maintenance stage in the project. Thus, the study has proposed a novel data format structure which can effectively link the 3D/BIM object with the maintenance data. The demonstration project shows how the data format structure is used. The contribution of this study is to provide guidance to a project practitioner by step-by-step approach in dealing with the significant maintenance information in the earlier stage of the construction project.
文摘According to the seismic and geological differences among every oil measures in mid-deep layers at west slope in Qikou Sag, varieties of new techniques on geophysics and geochemistry were introduced , such as seismic pro-cess of target and CWS, neural network seismic microfacies cluster analysis, 3D interval velocity analysis. Taking advan-tage of the series of new techniques aims at predicting synthetically and quantitatively the deep gravity flow channel sandbody. Furthermore, considering the structural setting data, potential structural-lithologic traps were specified. As a re-sult, the geological and drilling effect is obviously promoted.
基金The National Science and Technology Major Project of China under contract No.2011ZX05025-002-02the National Natural Science Foundation of China under contract Nos 41476032,91028009 and 40806019
文摘Turbidity currents represent a major agent for sediment transport in lakes, seas and oceans. In particu-lar, they formulate the most significant clastic accumulations in the deep sea, which become many of the world＇s most important hydrocarbon reservoirs. Several boreholes in the Qiongdongnan Basin, the north-western South China Sea, have recently revealed turbidity current deposits as significant hydrocarbon res-ervoirs. However, there are some arguments for the potential provenances. To solve this problem, it is es-sential to delineate their sedimentary processes as well as to evaluate their qualities as reservoir. Numerical simulations have been developed rapidly over the last several years, offering insights into turbidity current behaviors, as geologically significant turbidity currents are difficult to directly investigate due to their large scale and often destructive nature. Combined with the interpretation of the turbidity system based on high-resolution 3D seismic data, the paleotophography is acquired via a back-stripping seismic profile integrated with a borehole, i.e., Well A, in the western Qiongdongnan Basin; then a numerical model is built on the basis of this back-stripped profile. After defining the various turbidity current initial boundary conditions, includ-ing grain size, velocity and sediment concentration, the structures and behaviors of turbidity currents are investigated via numerical simulation software ANSYS FLUENT. Finally, the simulated turbidity deposits are compared with the interpreted sedimentary bodies based on 3D seismic data and the potential provenances of the revealed turbidites by Well A are discussed in details. The simulation results indicate that a sedimen-tary body develops far away from its source with an average grain size of 0.1 mm, i.e., sand-size sediment. Taking into account the location and orientation of the simulated seismic line, the consistence between normal forward simulation results and the revealed cores in Well A indicates that the turbidites should have been transported from Vietnam instead of Hainan Island. This interpretation has also been verified by the planar maps of sedimentary systems based on integration of boreholes and seismic data. The identification of the turbidity provenance will benefit the evaluation of extensively distributed submarine fans for hydro-carbon exploration in the deepwater areas.
基金supported by the National Natural Science Foundation of China(No.42072169)。
文摘Deep learning technologies are increasingly used in the fi eld of geophysics,and a variety of algorithms based on shallow convolutional neural networks are more widely used in fault recognition,but these methods are usually not able to accurately identify complex faults.In this study,using the advantage of deep residual networks to capture strong learning features,we introduce residual blocks to replace all convolutional layers of the three-dimensional(3D)UNet to build a new 3D Res-UNet and select appropriate parameters through experiments to train a large amount of synthesized seismic data.After the training is completed,we introduce the mechanism of knowledge distillation.First,we treat the 3D Res-UNet as a teacher network and then train the 3D Res-UNet as a student network;in this process,the teacher network is in evaluation mode.Finally,we calculate the mixed loss function by combining the teacher model and student network to learn more fault information,improve the performance of the network,and optimize the fault recognition eff ect.The quantitative evaluation result of the synthetic model test proves that the 3D Res-UNet can considerably improve the accuracy of fault recognition from 0.956 to 0.993 after knowledge distillation,and the eff ectiveness and feasibility of our method can be verifi ed based on the application of actual seismic data.
