By using a large amount of geological and geophysical data, the geological characteristics such as lithofacies and paleogeography of 4981 geological units at thirteen key geological periods or epoches since the Precam...By using a large amount of geological and geophysical data, the geological characteristics such as lithofacies and paleogeography of 4981 geological units at thirteen key geological periods or epoches since the Precambrian in the world have been figured out. The global lithofacies and paleogeography charts have been compiled by ArcGis mapping technology. Combined with the results of plate-paleogeography reconstruction, the lithofacies and paleogeography as well as the prototype basins of these global paleoplates have been restored with the Gplate software. Results show that there are 22 kinds of lithofacies combinations and 10 types of paleogeography units developed since Precambrian. These features of lithofacies and paleogeography as well as their evolution were mainly controlled by the divergent and convergent movements of those plates. Taking the results of the lithofacis and paleogeography at the present and paleoplate location during the seven key geological periods from the Precambrian to Paleozoic for example, during the Late Precambrian and Cambrian, the large-scale disintegration of the Rodinia supercontinent resulted in reduction of uplift denudation area and clastic terrestrial facies area, the expansion of coastal-shallow marine facies and shallow-water carbonate platform. In Devonian, uplift denudation area and clastic terrestrial facies area began to increase and littoral-shallow marine facies area and shallow-water carbonate platform shrank as a result of the formation of Larussia supercontinent. In the Permian, with the formation of the Pangea continent, the development of the global uplift denudation area and clastic terrestrial facies reached its peak, while the littoral and shallow marine facies were very limited in distribution. The lithofacies and paleogeography features and evolution patterns of different stages lay a solid foundation for analyzing the formation conditions of geological elements, such as source rocks, reservoirs and cap rocks for oil and gas accumulation, and revealing the distribution regularity of oil and gas around the world.展开更多
Recently,machine learning(ML)has been considered a powerful technological element of different society areas.To transform the computer into a decision maker,several sophisticated methods and algorithms are constantly ...Recently,machine learning(ML)has been considered a powerful technological element of different society areas.To transform the computer into a decision maker,several sophisticated methods and algorithms are constantly created and analyzed.In geophysics,both supervised and unsupervised ML methods have dramatically contributed to the development of seismic and well-log data interpretation.In well-logging,ML algorithms are well-suited for lithologic reconstruction problems,once there is no analytical expressions for computing well-log data produced by a particular rock unit.Additionally,supervised ML methods are strongly dependent on a accurate-labeled training data-set,which is not a simple task to achieve,due to data absences or corruption.Once an adequate supervision is performed,the classification outputs tend to be more accurate than unsupervised methods.This work presents a supervised version of a Self-Organizing Map,named as SSOM,to solve a lithologic reconstruction problem from well-log data.Firstly,we go for a more controlled problem and simulate well-log data directly from an interpreted geologic cross-section.We then define two specific training data-sets composed by density(RHOB),sonic(DT),spontaneous potential(SP)and gamma-ray(GR)logs,all simulated through a Gaussian distribution function per lithology.Once the training data-set is created,we simulate a particular pseudo-well,referred to as classification well,for defining controlled tests.First one comprises a training data-set with no labeled log data of the simulated fault zone.In the second test,we intentionally improve the training data-set with the fault.To bespeak the obtained results for each test,we analyze confusion matrices,logplots,accuracy and precision.Apart from very thin layer misclassifications,the SSOM provides reasonable lithologic reconstructions,especially when the improved training data-set is considered for supervision.The set of numerical experiments shows that our SSOM is extremely well-suited for a supervised lithologic reconstruction,especially to recover lithotypes that are weakly-sampled in the training log-data.On the other hand,some misclassifications are also observed when the cortex could not group the slightly different lithologies.展开更多
基金Supported by the China National Science and Technology Major Project(2011ZX05028-003,2016ZX05029-001)
文摘By using a large amount of geological and geophysical data, the geological characteristics such as lithofacies and paleogeography of 4981 geological units at thirteen key geological periods or epoches since the Precambrian in the world have been figured out. The global lithofacies and paleogeography charts have been compiled by ArcGis mapping technology. Combined with the results of plate-paleogeography reconstruction, the lithofacies and paleogeography as well as the prototype basins of these global paleoplates have been restored with the Gplate software. Results show that there are 22 kinds of lithofacies combinations and 10 types of paleogeography units developed since Precambrian. These features of lithofacies and paleogeography as well as their evolution were mainly controlled by the divergent and convergent movements of those plates. Taking the results of the lithofacis and paleogeography at the present and paleoplate location during the seven key geological periods from the Precambrian to Paleozoic for example, during the Late Precambrian and Cambrian, the large-scale disintegration of the Rodinia supercontinent resulted in reduction of uplift denudation area and clastic terrestrial facies area, the expansion of coastal-shallow marine facies and shallow-water carbonate platform. In Devonian, uplift denudation area and clastic terrestrial facies area began to increase and littoral-shallow marine facies area and shallow-water carbonate platform shrank as a result of the formation of Larussia supercontinent. In the Permian, with the formation of the Pangea continent, the development of the global uplift denudation area and clastic terrestrial facies reached its peak, while the littoral and shallow marine facies were very limited in distribution. The lithofacies and paleogeography features and evolution patterns of different stages lay a solid foundation for analyzing the formation conditions of geological elements, such as source rocks, reservoirs and cap rocks for oil and gas accumulation, and revealing the distribution regularity of oil and gas around the world.
文摘Recently,machine learning(ML)has been considered a powerful technological element of different society areas.To transform the computer into a decision maker,several sophisticated methods and algorithms are constantly created and analyzed.In geophysics,both supervised and unsupervised ML methods have dramatically contributed to the development of seismic and well-log data interpretation.In well-logging,ML algorithms are well-suited for lithologic reconstruction problems,once there is no analytical expressions for computing well-log data produced by a particular rock unit.Additionally,supervised ML methods are strongly dependent on a accurate-labeled training data-set,which is not a simple task to achieve,due to data absences or corruption.Once an adequate supervision is performed,the classification outputs tend to be more accurate than unsupervised methods.This work presents a supervised version of a Self-Organizing Map,named as SSOM,to solve a lithologic reconstruction problem from well-log data.Firstly,we go for a more controlled problem and simulate well-log data directly from an interpreted geologic cross-section.We then define two specific training data-sets composed by density(RHOB),sonic(DT),spontaneous potential(SP)and gamma-ray(GR)logs,all simulated through a Gaussian distribution function per lithology.Once the training data-set is created,we simulate a particular pseudo-well,referred to as classification well,for defining controlled tests.First one comprises a training data-set with no labeled log data of the simulated fault zone.In the second test,we intentionally improve the training data-set with the fault.To bespeak the obtained results for each test,we analyze confusion matrices,logplots,accuracy and precision.Apart from very thin layer misclassifications,the SSOM provides reasonable lithologic reconstructions,especially when the improved training data-set is considered for supervision.The set of numerical experiments shows that our SSOM is extremely well-suited for a supervised lithologic reconstruction,especially to recover lithotypes that are weakly-sampled in the training log-data.On the other hand,some misclassifications are also observed when the cortex could not group the slightly different lithologies.
