Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Ou...Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.展开更多
The working area is located in the industrially developed region of Rongshengpu-Qianjin, where a surface water system is developed, surface-layer lithology is complicated, and various kinds of hydrocarbon traps are bu...The working area is located in the industrially developed region of Rongshengpu-Qianjin, where a surface water system is developed, surface-layer lithology is complicated, and various kinds of hydrocarbon traps are buried at depth. The seismic data acquired previously couldn＇t be interpreted due to the complex surface and geological conditions. Taking secondary 3D seismic from the Rongshengpu-Qianjin area as an example, this paper describes a set of techniques designed to overcome these difficulties and improve the quality of seismic data. The applied techniques included flexible acquisition geometry, low-noise receiver conditions, quantitative quality control, and so on.展开更多
Forecasting subtle traps by sequence stratigraphy and 3D seismic data is a sensitive topic in hydrocarbon exploration. Research on subtle traps by geophysical data is the most popular and difficult. Based on the suffi...Forecasting subtle traps by sequence stratigraphy and 3D seismic data is a sensitive topic in hydrocarbon exploration. Research on subtle traps by geophysical data is the most popular and difficult. Based on the sufficiently drilling data, log data, core data and 3D seismic data, sediment sequence of Qikou depression, Huanghua basin was partitioned by using sequence stratigraphy theory. Each sediment sequence system mode was built. Sediment faces of subtle traps were pointed out. Dominating factors forming subtle traps were analyzed. Sandstone seismic rock physics and its response were studied in Tertiary System. Sandstone geophysical response and elastic modulus vary laws with pressure, temperature, porosity, depth were built. Experimental result and practice shows that it is possible using seismic information forecasting subtle traps. Integrated using geology, log, drilling data, special seismic processing technique, interpretation technique, high precision horizon calibration technique, 3D seismic visualizing interpretation, seismic coherence analysis, attribute analysis, logging-constrained inversion, time frequency analysis, subtle trapsobject is identified and interpreted. Finally, advantage object of subtle trap in this area was determined. Bottomland sand stratigraphic and lithologic reservoirs in Qinan slope zone have been founded by means of high resolution 3D seismic data field technique, high resolution 3D seismic data processing technique and seismic wave impendence inversion technique.展开更多
With the increase in mining depth,traditional coal mining methods not only waste coal resources but also seriously impact the stability of the roadway support structure during the collapse of the overburden rock.In co...With the increase in mining depth,traditional coal mining methods not only waste coal resources but also seriously impact the stability of the roadway support structure during the collapse of the overburden rock.In contrast,the top-cutting and depressurization technology utilizes the expansion effect of the rock effectively.This technology allows the rock body to collapse entirely,filling up the mining area through active intervention,which reduces the subsidence height of the overburden rock and significantly improves the coal extraction rate in the mining area.This study utilizes 3D seismic exploration technology to analyze the spatial distribution characteristics of fissure zones and rich zones of the rock strata in the mining area and investigate the movement law of overburdened rock during the coal seam mining process using the 110 mining method.It conducts numerical analysis combined with geomechanical modeling experiments to explore the movement law of the overburden rock under the influence of mining activities at Yuwang Coal Mine.The numerical analysis results indicate that the horizontal and vertical displacements of the rock body on the roof of the roadway are minimal when the angle of the slit is 75°.The overlying rock movement during the test is categorized by modeling the stress and strain fields into the following stages:fracture zone expansion,collapse zone gestation,rapid collapse zone development,and overlying rock stabilization.The rock on the cut side collapses more completely,breaking up and expanding to support the overburden,effectively reducing the depth of crack expansion and the extent of rock settlement and deformation.The integrity of the roadway roof remains intact during the rock collapse under NPR anchors.This study provides a scientific basis for understanding the movement law of overlying rock and for controlling the stability of the roadway perimeter rock in kilometer-deep underground mining.展开更多
文摘Deep learning, especially through convolutional neural networks (CNN) such as the U-Net 3D model, has revolutionized fault identification from seismic data, representing a significant leap over traditional methods. Our review traces the evolution of CNN, emphasizing the adaptation and capabilities of the U-Net 3D model in automating seismic fault delineation with unprecedented accuracy. We find: 1) The transition from basic neural networks to sophisticated CNN has enabled remarkable advancements in image recognition, which are directly applicable to analyzing seismic data. The U-Net 3D model, with its innovative architecture, exemplifies this progress by providing a method for detailed and accurate fault detection with reduced manual interpretation bias. 2) The U-Net 3D model has demonstrated its superiority over traditional fault identification methods in several key areas: it has enhanced interpretation accuracy, increased operational efficiency, and reduced the subjectivity of manual methods. 3) Despite these achievements, challenges such as the need for effective data preprocessing, acquisition of high-quality annotated datasets, and achieving model generalization across different geological conditions remain. Future research should therefore focus on developing more complex network architectures and innovative training strategies to refine fault identification performance further. Our findings confirm the transformative potential of deep learning, particularly CNN like the U-Net 3D model, in geosciences, advocating for its broader integration to revolutionize geological exploration and seismic analysis.
文摘The working area is located in the industrially developed region of Rongshengpu-Qianjin, where a surface water system is developed, surface-layer lithology is complicated, and various kinds of hydrocarbon traps are buried at depth. The seismic data acquired previously couldn＇t be interpreted due to the complex surface and geological conditions. Taking secondary 3D seismic from the Rongshengpu-Qianjin area as an example, this paper describes a set of techniques designed to overcome these difficulties and improve the quality of seismic data. The applied techniques included flexible acquisition geometry, low-noise receiver conditions, quantitative quality control, and so on.
基金Project(2003034470) supported by the Postdoctoral Science Foundation of China project supported by the Postdoctoral Science Foundation of Central South University
文摘Forecasting subtle traps by sequence stratigraphy and 3D seismic data is a sensitive topic in hydrocarbon exploration. Research on subtle traps by geophysical data is the most popular and difficult. Based on the sufficiently drilling data, log data, core data and 3D seismic data, sediment sequence of Qikou depression, Huanghua basin was partitioned by using sequence stratigraphy theory. Each sediment sequence system mode was built. Sediment faces of subtle traps were pointed out. Dominating factors forming subtle traps were analyzed. Sandstone seismic rock physics and its response were studied in Tertiary System. Sandstone geophysical response and elastic modulus vary laws with pressure, temperature, porosity, depth were built. Experimental result and practice shows that it is possible using seismic information forecasting subtle traps. Integrated using geology, log, drilling data, special seismic processing technique, interpretation technique, high precision horizon calibration technique, 3D seismic visualizing interpretation, seismic coherence analysis, attribute analysis, logging-constrained inversion, time frequency analysis, subtle trapsobject is identified and interpreted. Finally, advantage object of subtle trap in this area was determined. Bottomland sand stratigraphic and lithologic reservoirs in Qinan slope zone have been founded by means of high resolution 3D seismic data field technique, high resolution 3D seismic data processing technique and seismic wave impendence inversion technique.
基金Study on in-situ stress database and 3D in-situ stress inversion technology of highway tunnel in Shanxi Province(22-JKCF-08)the Study on disaster mechanism and NPR anchor cable prevention and control of coal mining caving subsidence in operating tunnel in mountainous area(2022-JKKJ-6)for their support+2 种基金supported by the State Key Laboratory for Geomechanics and Deep Underground Engineering(SKLGDUEK2020)Huaneng Group headquarters science and technology project(HNKJ21-H07the Coal Burst Research Center of China Jiangsu.
文摘With the increase in mining depth,traditional coal mining methods not only waste coal resources but also seriously impact the stability of the roadway support structure during the collapse of the overburden rock.In contrast,the top-cutting and depressurization technology utilizes the expansion effect of the rock effectively.This technology allows the rock body to collapse entirely,filling up the mining area through active intervention,which reduces the subsidence height of the overburden rock and significantly improves the coal extraction rate in the mining area.This study utilizes 3D seismic exploration technology to analyze the spatial distribution characteristics of fissure zones and rich zones of the rock strata in the mining area and investigate the movement law of overburdened rock during the coal seam mining process using the 110 mining method.It conducts numerical analysis combined with geomechanical modeling experiments to explore the movement law of the overburden rock under the influence of mining activities at Yuwang Coal Mine.The numerical analysis results indicate that the horizontal and vertical displacements of the rock body on the roof of the roadway are minimal when the angle of the slit is 75°.The overlying rock movement during the test is categorized by modeling the stress and strain fields into the following stages:fracture zone expansion,collapse zone gestation,rapid collapse zone development,and overlying rock stabilization.The rock on the cut side collapses more completely,breaking up and expanding to support the overburden,effectively reducing the depth of crack expansion and the extent of rock settlement and deformation.The integrity of the roadway roof remains intact during the rock collapse under NPR anchors.This study provides a scientific basis for understanding the movement law of overlying rock and for controlling the stability of the roadway perimeter rock in kilometer-deep underground mining.
