The increasingly rapid development of the disciplines of petroleum engineering and petroleum geology has led to new methodologies and interpretation techniques forming new knowledge that should be offered quickly and ...The increasingly rapid development of the disciplines of petroleum engineering and petroleum geology has led to new methodologies and interpretation techniques forming new knowledge that should be offered quickly and efficiently to modern engineers and geologists. This need is equally important for students as well as for young professionals. Access and training to all scientific information is necessary to ensure success in their future careers. Today, e-learning has become a common medium for the management and distribution of on-line educational content. Learning Management Systems （LMSs） were not only developed to handle a large variety of multimedia content that provides an organized knowledge repository used to accelerate access to information and skill acquisition; but, LMSs can also keep detailed statistics on the use of the available material offering a powerful training and educational tool. In this document, the Petroleum Knowledge Tutorial System, an LMS platform offering a variety of online educational and training options to petroleum engineers and geologists, is presented. It was created using Moodle, open- source software that can be used to create on-line courses. The platform covers fundamental educational concepts in a structured way. It follows an optimized ＂workflow＂ that can be applied not only to solve a specific exercise but also any similar problem encountered over the course of one＇s career. The platform was designed to offer a repository of learning material in various forms and to favor user-platform interactions. It can be used for training and evaluation purposes through exercises and problem solving that the user can perform online by using browsing software along with internet access. Special tools were created and implemented on the platform to assist the user in completing a variety of tasks including performing exercises involving calculations with given data and plots of points or lines on graphs without leaving the learning environment. Furthermore, videos with detailed explanations follow each learning module and provide the full solution to every exercise. The LMS automatically keeps a large statistical database including the users＇ access to activities on the platform that can be exported and further processed to improve the platform functionality and evaluate the users＇ performance.展开更多
With the rapid development of deep learning neural networks,new solutions have emerged for addressing fluid flow problems in porous media.Combining data-driven approaches with physical constraints has become a hot res...With the rapid development of deep learning neural networks,new solutions have emerged for addressing fluid flow problems in porous media.Combining data-driven approaches with physical constraints has become a hot research direction,with physics-informed neural networks(PINNs) being the most popular hybrid model.PINNs have gained widespread attention in subsurface fluid flow simulations due to their low computational resource requirements,fast training speeds,strong generalization capabilities,and broad applicability.Despite success in homogeneous settings,standard PINNs face challenges in accurately calculating flux between irregular Eulerian cells with disparate properties and capturing global field influences on local cells.This limits their suitability for heterogeneous reservoirs and the irregular Eulerian grids frequently used in reservoir.To address these challenges,this study proposes a physics-informed graph neural network(PIGNN) model.The PIGNN model treats the entire field as a whole,integrating information from neighboring grids and physical laws into the solution for the target grid,thereby improving the accuracy of solving partial differential equations in heterogeneous and Eulerian irregular grids.The optimized model was applied to pressure field prediction in a spatially heterogeneous reservoir,achieving an average L_(2) error and R_(2) score of 6.710×10^(-4)and 0.998,respectively,which confirms the effectiveness of model.Compared to the conventional PINN model,the average L_(2) error was reduced by 76.93%,the average R_(2) score increased by 3.56%.Moreover,evaluating robustness,training the PIGNN model using only 54% and 76% of the original data yielded average relative L_(2) error reductions of 58.63% and 56.22%,respectively,compared to the PINN model.These results confirm the superior performance of this approach compared to PINN.展开更多
Due to the high viscosity of heavy crude oils,production from these reservoirs is a demanding task.To tackle this problem,reducing oil viscosity is a promising approach.There are various methods to reduce viscosity of...Due to the high viscosity of heavy crude oils,production from these reservoirs is a demanding task.To tackle this problem,reducing oil viscosity is a promising approach.There are various methods to reduce viscosity of heavy oil:heating,diluting,emulsification,and core annular flow.In this study,dilution approach was employed,using industrial solvents and gas condensate.The viscosity of two Iranian heavy crude oils was measured by mixing with solvents at different temperatures.Dilution of both oil samples with toluene and heptane,resulted in viscosity reduction.However,their effect became less significant at higher concentrations of diluent.Because of forming hydrogen bonds,adding methanol to heavy crude oil resulted in higher viscosity.By adding condensate,viscosity of each sample reduced.Gas condensate had a greater impact on heavier oil;however,at higher temperatures its effect was reduced.Diluting with naphtha decreased heavy oil viscosity in the same way as n-heptane and toluene.Besides experimental investigation,different viscosity models were evaluated for prediction of heavy oil/solvent viscosity.It was recognized that Lederer'model is the best one.展开更多
Density is an important property of natural gas required for the design of gas processing and reservoir simulation.Due to expensive measurement of density,industry tends to predict gas density through an EOS.However,a...Density is an important property of natural gas required for the design of gas processing and reservoir simulation.Due to expensive measurement of density,industry tends to predict gas density through an EOS.However,all EOS are associated with uncertainties,especially at highpressure conditions.Also,using sophisticated EOS in commercial software renders simulation highly time-consuming.This work aims to evaluate performance of adaptive neuro-fuzzy inference system(ANFIS)as a widely-accepted intelligent model for prediction of P-r-T behavior of natural gas.Using experimental data reported in the literature,our inference system was trained with 95 data of natural gas densities in the temperature range of(250-450)K and pressures up to 150 MPa.Additionally,prediction by ANFIS was compared with those of AGA8 and GERG04 which both are leading industrial EOS for calculation of natural gas density.It was observed that ANFIS predicts natural gas density with AARD%of 1.704;and is able to estimate gas density as accurate as sophisticated EOS.The proposed model is applicable for predicting gas density in the range of(250-450)K,(10-150)MPa and also for sweet gases,i.e.,containing a low concentration of N2 and CO2.展开更多
The understanding of the mechanisms behind foam generation and the structure of foam itself form the basis of foam-related experiments for its application in Enhanced Oil Recovery and overcoming gas in-jection limitat...The understanding of the mechanisms behind foam generation and the structure of foam itself form the basis of foam-related experiments for its application in Enhanced Oil Recovery and overcoming gas in-jection limitations.Novel insights in this paper towards the theory of foam generation can help explain experimental results and lead to improved formulas of the applied substances and concentrations.This study aims to investigate the mechanisms behind foam generation and the structure of foam by specific laboratory experiments and theoretical analyses.The liquid drainage through interconnected Plateau borders was found to be the most critical foam decay mechanism for this particular research.The justification of the foam drainage equation was demonstrated by comparing the numerical solution with the outcome of a few bulk experiments.The discrepancies were described according to the limitations of both the theory and the experimental settings.Foam modelling gives more profound knowledge in more detail of the different stages in foam drainage than experimental data can deliver,which is because of the lack of continuous measurement of foam conductivity for the foam bulk test.Therefore,a comprehension of foam modelling investigation and comparison is required to gain a deeper understanding of foam behaviour.展开更多
文摘The increasingly rapid development of the disciplines of petroleum engineering and petroleum geology has led to new methodologies and interpretation techniques forming new knowledge that should be offered quickly and efficiently to modern engineers and geologists. This need is equally important for students as well as for young professionals. Access and training to all scientific information is necessary to ensure success in their future careers. Today, e-learning has become a common medium for the management and distribution of on-line educational content. Learning Management Systems （LMSs） were not only developed to handle a large variety of multimedia content that provides an organized knowledge repository used to accelerate access to information and skill acquisition; but, LMSs can also keep detailed statistics on the use of the available material offering a powerful training and educational tool. In this document, the Petroleum Knowledge Tutorial System, an LMS platform offering a variety of online educational and training options to petroleum engineers and geologists, is presented. It was created using Moodle, open- source software that can be used to create on-line courses. The platform covers fundamental educational concepts in a structured way. It follows an optimized ＂workflow＂ that can be applied not only to solve a specific exercise but also any similar problem encountered over the course of one＇s career. The platform was designed to offer a repository of learning material in various forms and to favor user-platform interactions. It can be used for training and evaluation purposes through exercises and problem solving that the user can perform online by using browsing software along with internet access. Special tools were created and implemented on the platform to assist the user in completing a variety of tasks including performing exercises involving calculations with given data and plots of points or lines on graphs without leaving the learning environment. Furthermore, videos with detailed explanations follow each learning module and provide the full solution to every exercise. The LMS automatically keeps a large statistical database including the users＇ access to activities on the platform that can be exported and further processed to improve the platform functionality and evaluate the users＇ performance.
基金supported by the National Natural Science Foundation of China (No. 52274048)Beijing Natural Science Foundation (No. 3222037)。
文摘With the rapid development of deep learning neural networks,new solutions have emerged for addressing fluid flow problems in porous media.Combining data-driven approaches with physical constraints has become a hot research direction,with physics-informed neural networks(PINNs) being the most popular hybrid model.PINNs have gained widespread attention in subsurface fluid flow simulations due to their low computational resource requirements,fast training speeds,strong generalization capabilities,and broad applicability.Despite success in homogeneous settings,standard PINNs face challenges in accurately calculating flux between irregular Eulerian cells with disparate properties and capturing global field influences on local cells.This limits their suitability for heterogeneous reservoirs and the irregular Eulerian grids frequently used in reservoir.To address these challenges,this study proposes a physics-informed graph neural network(PIGNN) model.The PIGNN model treats the entire field as a whole,integrating information from neighboring grids and physical laws into the solution for the target grid,thereby improving the accuracy of solving partial differential equations in heterogeneous and Eulerian irregular grids.The optimized model was applied to pressure field prediction in a spatially heterogeneous reservoir,achieving an average L_(2) error and R_(2) score of 6.710×10^(-4)and 0.998,respectively,which confirms the effectiveness of model.Compared to the conventional PINN model,the average L_(2) error was reduced by 76.93%,the average R_(2) score increased by 3.56%.Moreover,evaluating robustness,training the PIGNN model using only 54% and 76% of the original data yielded average relative L_(2) error reductions of 58.63% and 56.22%,respectively,compared to the PINN model.These results confirm the superior performance of this approach compared to PINN.
文摘Due to the high viscosity of heavy crude oils,production from these reservoirs is a demanding task.To tackle this problem,reducing oil viscosity is a promising approach.There are various methods to reduce viscosity of heavy oil:heating,diluting,emulsification,and core annular flow.In this study,dilution approach was employed,using industrial solvents and gas condensate.The viscosity of two Iranian heavy crude oils was measured by mixing with solvents at different temperatures.Dilution of both oil samples with toluene and heptane,resulted in viscosity reduction.However,their effect became less significant at higher concentrations of diluent.Because of forming hydrogen bonds,adding methanol to heavy crude oil resulted in higher viscosity.By adding condensate,viscosity of each sample reduced.Gas condensate had a greater impact on heavier oil;however,at higher temperatures its effect was reduced.Diluting with naphtha decreased heavy oil viscosity in the same way as n-heptane and toluene.Besides experimental investigation,different viscosity models were evaluated for prediction of heavy oil/solvent viscosity.It was recognized that Lederer'model is the best one.
文摘Density is an important property of natural gas required for the design of gas processing and reservoir simulation.Due to expensive measurement of density,industry tends to predict gas density through an EOS.However,all EOS are associated with uncertainties,especially at highpressure conditions.Also,using sophisticated EOS in commercial software renders simulation highly time-consuming.This work aims to evaluate performance of adaptive neuro-fuzzy inference system(ANFIS)as a widely-accepted intelligent model for prediction of P-r-T behavior of natural gas.Using experimental data reported in the literature,our inference system was trained with 95 data of natural gas densities in the temperature range of(250-450)K and pressures up to 150 MPa.Additionally,prediction by ANFIS was compared with those of AGA8 and GERG04 which both are leading industrial EOS for calculation of natural gas density.It was observed that ANFIS predicts natural gas density with AARD%of 1.704;and is able to estimate gas density as accurate as sophisticated EOS.The proposed model is applicable for predicting gas density in the range of(250-450)K,(10-150)MPa and also for sweet gases,i.e.,containing a low concentration of N2 and CO2.
文摘The understanding of the mechanisms behind foam generation and the structure of foam itself form the basis of foam-related experiments for its application in Enhanced Oil Recovery and overcoming gas in-jection limitations.Novel insights in this paper towards the theory of foam generation can help explain experimental results and lead to improved formulas of the applied substances and concentrations.This study aims to investigate the mechanisms behind foam generation and the structure of foam by specific laboratory experiments and theoretical analyses.The liquid drainage through interconnected Plateau borders was found to be the most critical foam decay mechanism for this particular research.The justification of the foam drainage equation was demonstrated by comparing the numerical solution with the outcome of a few bulk experiments.The discrepancies were described according to the limitations of both the theory and the experimental settings.Foam modelling gives more profound knowledge in more detail of the different stages in foam drainage than experimental data can deliver,which is because of the lack of continuous measurement of foam conductivity for the foam bulk test.Therefore,a comprehension of foam modelling investigation and comparison is required to gain a deeper understanding of foam behaviour.
