Flowing bottom-hole pressure(FBHP)is a key metric parameter in the evaluation of performances of oil and gas production wells.An accurate prediction of FBHP is highly required in the petroleum industry for many applic...Flowing bottom-hole pressure(FBHP)is a key metric parameter in the evaluation of performances of oil and gas production wells.An accurate prediction of FBHP is highly required in the petroleum industry for many applications,such the hydrocarbon production optimization,oil lifting cost,and assessment of workover operations.Production and reservoir engineers rely on empirical correlations and mechanistic models exist in open resources to estimate the FBHP.Several empirical models have been developed based on simulation and laboratory results that involved many assumptions that reduce the model's accuracy when they are applied for the field applications.The technologies of machine learning(ML)are one discipline of Artificial Intelligence(AI)techniques provide promising tools that help solving human's complex problems.This study develops machine-learning based models to predict the multiphase FBHP using three machine learning techniques that are Random forest,K-Nearest Neighbors(KNN),and artificial neural network(ANN).Results showed that using an artificial neural network model give error of 2.5%to estimate the FBHP which is less than the random forest and K-nearest neighbor models with error of 3.6%and 4%respectively.The ML models were developed based on a surface production data,which makes the FBHP is predicted using actual field data.The accuracy of the proposed models from ML was evaluated by comparing the results with the actual dataset values to ensure the effectiveness of the work.The results of this study show the potential of artificial intelligence in predicting the most complex parameter in the multiphase petroleum production process.展开更多
To design an efficient intermittent gas-lift installation,reliable information is needed in the performance of all process components,from the outer boundary of the reservoir to the surface separators.The gas lift val...To design an efficient intermittent gas-lift installation,reliable information is needed in the performance of all process components,from the outer boundary of the reservoir to the surface separators.The gas lift valve is the one critical component that affects the design of the whole system.In intermittent producing system,the pilot gas-lift valve is extremely used to control the point of compressed gas entry into the production tubing and acts as a pressure regulator.A novel approach using computational fluid dynamics simulation was performed in this study to develop a dynamic model for the gas passage performance of a 1-in.,Nitrogen-charged,pilot gas-lift valve.Dynamic performance curves were obtained by using Methane as an injection gas with flow rates reaching up to 4.5 MMscf/day.This study investigates the effect of internal pressure,velocity and temperature distribution within the pilot valve that cannot be predicted in the experiments and mathematical models during the flow-performance studies.A general equation of the nonconstant discharge coefficient has been developed for 1-inch pilot valve to be used for further calculation in the industry without using CFD model.The developed model significantly reduces the complexity of the data required to calculate the discharge coefficient.展开更多
A computational fluid dynamics model(CFD)is developed for intermittent gas lift techniques.The simulation is conducted for a test section of 18 m vertical tube with 0.076 m in diameter using air as injection gas and o...A computational fluid dynamics model(CFD)is developed for intermittent gas lift techniques.The simulation is conducted for a test section of 18 m vertical tube with 0.076 m in diameter using air as injection gas and oil as a formation fluid.The results obtained from the CFD model are validated with the experiment results from the literature.The current study shows that computational modeling is a proven simulation program for predicting intermittent gas lift characteristics and the transient flow parameters that are changing with time and position in the coordinate system.The model can predict the slug velocity behavior for different injection pressure.The slug velocity profile shows three regions;the first region is the rapid acceleration at the initial time of injection,the second region shows the nearly constant velocity until the slug reaches the surface and the third region is again the rapid acceleration when the liquid starts to produce.Also,the results obtained from this model show that as the gas injection pressure increases,the liquid slug velocity increase,and the region of the constant velocity decrease.The effect of the injection time on the liquid production rate has been studied for two different gas injection pressures of 40 psig and 50 psig.The developed model shows that more than 50%of the liquid production is coming from after flow period.展开更多
Tubing pressure at gas injection depth in intermittent wells is one of the most critical parameters for production engineers to evaluate the performance of the system.However,monitoring of the tubing pressure is not u...Tubing pressure at gas injection depth in intermittent wells is one of the most critical parameters for production engineers to evaluate the performance of the system.However,monitoring of the tubing pressure is not usually carried out in real time.It has been realized that the generally used correlations are not effective enough due to complexity of the intermittent process which involve many parameters and assumptions to develop such equations.The focus of this study is to utilize machine learning(ML)algorithms to develop a model that can accurately predict tubing pressure in artificial intermittent gas lift wells.intelligent algorithms built on the field data provide a solution that is easy to use and universally applicable to the complex problems.Various non-linear regression ML methods are employed in this study,namely,Decision Tree-regression(DT),Random Forest-regression(RF)and K Nearest Neighbors-regression(KNN).All the tubing pressures obtained from ML models were compared with the actual values to ensure the effectiveness of the work.The developed models show that it can predict the pressure with more than 99.9%accuracy.This is an interesting result,as such outcome accuracy has not been reported usually in the open literature.展开更多
文摘Flowing bottom-hole pressure(FBHP)is a key metric parameter in the evaluation of performances of oil and gas production wells.An accurate prediction of FBHP is highly required in the petroleum industry for many applications,such the hydrocarbon production optimization,oil lifting cost,and assessment of workover operations.Production and reservoir engineers rely on empirical correlations and mechanistic models exist in open resources to estimate the FBHP.Several empirical models have been developed based on simulation and laboratory results that involved many assumptions that reduce the model's accuracy when they are applied for the field applications.The technologies of machine learning(ML)are one discipline of Artificial Intelligence(AI)techniques provide promising tools that help solving human's complex problems.This study develops machine-learning based models to predict the multiphase FBHP using three machine learning techniques that are Random forest,K-Nearest Neighbors(KNN),and artificial neural network(ANN).Results showed that using an artificial neural network model give error of 2.5%to estimate the FBHP which is less than the random forest and K-nearest neighbor models with error of 3.6%and 4%respectively.The ML models were developed based on a surface production data,which makes the FBHP is predicted using actual field data.The accuracy of the proposed models from ML was evaluated by comparing the results with the actual dataset values to ensure the effectiveness of the work.The results of this study show the potential of artificial intelligence in predicting the most complex parameter in the multiphase petroleum production process.
基金This study was carried out as part of the EFOP-3.6.1-16-2016-00011"Younger and Renewing University一Innovative Knowledge City-institutional development of the University of Miskolc aiming at intelligent specialization"project implemented in the framework of the Szechenyi 2020 program.The realization of this project is supported by the European Union,co-financed by the European Social Fund.
文摘To design an efficient intermittent gas-lift installation,reliable information is needed in the performance of all process components,from the outer boundary of the reservoir to the surface separators.The gas lift valve is the one critical component that affects the design of the whole system.In intermittent producing system,the pilot gas-lift valve is extremely used to control the point of compressed gas entry into the production tubing and acts as a pressure regulator.A novel approach using computational fluid dynamics simulation was performed in this study to develop a dynamic model for the gas passage performance of a 1-in.,Nitrogen-charged,pilot gas-lift valve.Dynamic performance curves were obtained by using Methane as an injection gas with flow rates reaching up to 4.5 MMscf/day.This study investigates the effect of internal pressure,velocity and temperature distribution within the pilot valve that cannot be predicted in the experiments and mathematical models during the flow-performance studies.A general equation of the nonconstant discharge coefficient has been developed for 1-inch pilot valve to be used for further calculation in the industry without using CFD model.The developed model significantly reduces the complexity of the data required to calculate the discharge coefficient.
基金This study was carried out as part of the EFOP-3.6.1-16-2016-00011“Younger and Renewing University e Innovative Knowledge City e institutional development of the University of Miskolc aiming at intelligent specialisation”project implemented in the framework of the Szechenyi 2020 program,The realization of this project is supported by the European Union,co-financed by the European Social Fund.
文摘A computational fluid dynamics model(CFD)is developed for intermittent gas lift techniques.The simulation is conducted for a test section of 18 m vertical tube with 0.076 m in diameter using air as injection gas and oil as a formation fluid.The results obtained from the CFD model are validated with the experiment results from the literature.The current study shows that computational modeling is a proven simulation program for predicting intermittent gas lift characteristics and the transient flow parameters that are changing with time and position in the coordinate system.The model can predict the slug velocity behavior for different injection pressure.The slug velocity profile shows three regions;the first region is the rapid acceleration at the initial time of injection,the second region shows the nearly constant velocity until the slug reaches the surface and the third region is again the rapid acceleration when the liquid starts to produce.Also,the results obtained from this model show that as the gas injection pressure increases,the liquid slug velocity increase,and the region of the constant velocity decrease.The effect of the injection time on the liquid production rate has been studied for two different gas injection pressures of 40 psig and 50 psig.The developed model shows that more than 50%of the liquid production is coming from after flow period.
文摘Tubing pressure at gas injection depth in intermittent wells is one of the most critical parameters for production engineers to evaluate the performance of the system.However,monitoring of the tubing pressure is not usually carried out in real time.It has been realized that the generally used correlations are not effective enough due to complexity of the intermittent process which involve many parameters and assumptions to develop such equations.The focus of this study is to utilize machine learning(ML)algorithms to develop a model that can accurately predict tubing pressure in artificial intermittent gas lift wells.intelligent algorithms built on the field data provide a solution that is easy to use and universally applicable to the complex problems.Various non-linear regression ML methods are employed in this study,namely,Decision Tree-regression(DT),Random Forest-regression(RF)and K Nearest Neighbors-regression(KNN).All the tubing pressures obtained from ML models were compared with the actual values to ensure the effectiveness of the work.The developed models show that it can predict the pressure with more than 99.9%accuracy.This is an interesting result,as such outcome accuracy has not been reported usually in the open literature.