The total organic carbon content usually determines the hydrocarbon generation potential of a formation.A higher total organic carbon content often corresponds to a greater possibility of generating large amounts of o...The total organic carbon content usually determines the hydrocarbon generation potential of a formation.A higher total organic carbon content often corresponds to a greater possibility of generating large amounts of oil or gas.Hence,accurately calculating the total organic carbon content in a formation is very important.Present research is focused on precisely calculating the total organic carbon content based on machine learning.At present,many machine learning methods,including backpropagation neural networks,support vector regression,random forests,extreme learning machines,and deep learning,are employed to evaluate the total organic carbon content.However,the principles and perspectives of various machine learning algorithms are quite different.This paper reviews the application of various machine learning algorithms to deal with total organic carbon content evaluation problems.Of various machine learning algorithms used for TOC content predication,two algorithms,the backpropagation neural network and support vector regression are the most commonly used,and the backpropagation neural network is sometimes combined with many other algorithms to achieve better results.Additionally,combining multiple algorithms or using deep learning to increase the number of network layers can further improve the total organic carbon content prediction.The prediction by backpropagation neural network may be better than that by support vector regression;nevertheless,using any type of machine learning algorithm improves the total organic carbon content prediction in a given research block.According to some published literature,the determination coefficient(R^(2))can be increased by up to 0.46 after using machine learning.Deep learning algorithms may be the next breakthrough direction that can significantly improve the prediction of the total organic carbon content.Evaluating the total organic carbon content based on machine learning is of great significance.展开更多
Accurate quantification of the gas hydrate content in the deep sea is useful for assessing the resource potential and understanding the role of gas hydrates in the global carbon cycle.Resistivity logging data combined...Accurate quantification of the gas hydrate content in the deep sea is useful for assessing the resource potential and understanding the role of gas hydrates in the global carbon cycle.Resistivity logging data combined with Archie’s equation are often used to calculate gas hydrate saturation,but the reliability is dependent on the rationality of the empirical parameter cementation factor and saturation index.At present,an increasing number of fine-grained hydrate-rich sediment regions have been discovered worldwide through drilling efforts,and the reservoir types and hydrate distribution are diverse,which differs greatly from that of coarse-grained reservoirs of hydrate-bearing sediment.This results in vertical variations in m and n through stratigraphy.At present,the saturation evaluation effect of these reservoirs cannot be improved.In this work,a theory for the determination of the cementation factor and saturation index was first proposed to obtain reliable and variable values of the empirical parameters.Then,a hydrate saturation evaluation technique with variables m and n was formed based on the well logging data.This technique was used to evaluate complex fine-grained hydrate-bearing reservoirs in several regions worldwide.It was found that the highest n could be 16,and the log calculation results were more consistent with the core hydrate saturation.Additionally,the cause of the excessively high n values was explained from physical principles,and the result was verified with actually well log data.In future evaluations of the amount of hydrate resources in fine-grained sediment reservoirs worldwide,new saturation estimation methods should be taken into account to advance hydrate research.展开更多
The identification of high-quality marine shale gas reservoirs has always been a key task in the exploration and development stage.However,due to the serious nonlinear relationship between the logging curve response a...The identification of high-quality marine shale gas reservoirs has always been a key task in the exploration and development stage.However,due to the serious nonlinear relationship between the logging curve response and high-quality reservoirs,the rapid identification of high-quality reservoirs has always been a problem of low accuracy.This study proposes a combination of the oversampling method and random forest algorithm to improve the identification accuracy of high-quality reservoirs based on logging data.The oversampling method is used to balance the number of samples of different types and the random forest algorithm is used to establish a highprecision and high-quality reservoir identification model.From the perspective of the prediction effect,the reservoir identification method that combines the oversampling method and the random forest algorithm has increased the accuracy of reservoir identification from the 44%seen in other machine learning algorithms to 78%,and the effect is significant.This research can improve the identifiability of high-quality marine shale gas reservoirs,guide the drilling of horizontal wells,and provide tangible help for the precise formulation of marine shale gas development plans.展开更多
基金This project was funded by the Open Fund of the Key Laboratory of Exploration Technologies for Oil and Gas Resources,the Ministry of Education(No.K2021-03)National Natural Science Foundation of China(No.42106213)+2 种基金the Hainan Provincial Natural Science Foundation of China(No.421QN281)the China Postdoctoral Science Foundation(Nos.2021M690161 and 2021T140691)the Postdoctorate Funded Project in Hainan Province.
文摘The total organic carbon content usually determines the hydrocarbon generation potential of a formation.A higher total organic carbon content often corresponds to a greater possibility of generating large amounts of oil or gas.Hence,accurately calculating the total organic carbon content in a formation is very important.Present research is focused on precisely calculating the total organic carbon content based on machine learning.At present,many machine learning methods,including backpropagation neural networks,support vector regression,random forests,extreme learning machines,and deep learning,are employed to evaluate the total organic carbon content.However,the principles and perspectives of various machine learning algorithms are quite different.This paper reviews the application of various machine learning algorithms to deal with total organic carbon content evaluation problems.Of various machine learning algorithms used for TOC content predication,two algorithms,the backpropagation neural network and support vector regression are the most commonly used,and the backpropagation neural network is sometimes combined with many other algorithms to achieve better results.Additionally,combining multiple algorithms or using deep learning to increase the number of network layers can further improve the total organic carbon content prediction.The prediction by backpropagation neural network may be better than that by support vector regression;nevertheless,using any type of machine learning algorithm improves the total organic carbon content prediction in a given research block.According to some published literature,the determination coefficient(R^(2))can be increased by up to 0.46 after using machine learning.Deep learning algorithms may be the next breakthrough direction that can significantly improve the prediction of the total organic carbon content.Evaluating the total organic carbon content based on machine learning is of great significance.
基金This project was funded by the National Natural Science Foundation of China(No.42106213)the China Postdoctoral Science Foundation(Nos.2021M690161 and 2021T140691)+1 种基金the National Key Research and Development Program of China(No.2021YFC3100601)the Postdoctorate Funded Project in Hainan Province.
文摘Accurate quantification of the gas hydrate content in the deep sea is useful for assessing the resource potential and understanding the role of gas hydrates in the global carbon cycle.Resistivity logging data combined with Archie’s equation are often used to calculate gas hydrate saturation,but the reliability is dependent on the rationality of the empirical parameter cementation factor and saturation index.At present,an increasing number of fine-grained hydrate-rich sediment regions have been discovered worldwide through drilling efforts,and the reservoir types and hydrate distribution are diverse,which differs greatly from that of coarse-grained reservoirs of hydrate-bearing sediment.This results in vertical variations in m and n through stratigraphy.At present,the saturation evaluation effect of these reservoirs cannot be improved.In this work,a theory for the determination of the cementation factor and saturation index was first proposed to obtain reliable and variable values of the empirical parameters.Then,a hydrate saturation evaluation technique with variables m and n was formed based on the well logging data.This technique was used to evaluate complex fine-grained hydrate-bearing reservoirs in several regions worldwide.It was found that the highest n could be 16,and the log calculation results were more consistent with the core hydrate saturation.Additionally,the cause of the excessively high n values was explained from physical principles,and the result was verified with actually well log data.In future evaluations of the amount of hydrate resources in fine-grained sediment reservoirs worldwide,new saturation estimation methods should be taken into account to advance hydrate research.
基金This project was funded by the Laboratory for Marine Geology,Qingdao National Laboratory for Marine Science and Technology,(MGQNLM-KF202004)China Postdoctoral Science Foundation(2021M690161,2021T140691)+2 种基金Postdoctoral Funded Project in Hainan Province(General Program)Chinese Academy of Sciences-Special Research Assistant Projectthe Open Fund of Key Laboratory of Exploration Technologies for Oil and Gas Resources(Yangtze University),Ministry of Education(No.K2021–03,K2021-08)。
文摘The identification of high-quality marine shale gas reservoirs has always been a key task in the exploration and development stage.However,due to the serious nonlinear relationship between the logging curve response and high-quality reservoirs,the rapid identification of high-quality reservoirs has always been a problem of low accuracy.This study proposes a combination of the oversampling method and random forest algorithm to improve the identification accuracy of high-quality reservoirs based on logging data.The oversampling method is used to balance the number of samples of different types and the random forest algorithm is used to establish a highprecision and high-quality reservoir identification model.From the perspective of the prediction effect,the reservoir identification method that combines the oversampling method and the random forest algorithm has increased the accuracy of reservoir identification from the 44%seen in other machine learning algorithms to 78%,and the effect is significant.This research can improve the identifiability of high-quality marine shale gas reservoirs,guide the drilling of horizontal wells,and provide tangible help for the precise formulation of marine shale gas development plans.
