To achieve carbon dioxide(CO_(2))storage through enhanced oil recovery,accurate forecasting of CO_(2) subsurface storage and cumulative oil production is essential.This study develops hybrid predictive models for the ...To achieve carbon dioxide(CO_(2))storage through enhanced oil recovery,accurate forecasting of CO_(2) subsurface storage and cumulative oil production is essential.This study develops hybrid predictive models for the determination of CO_(2) storage mass and cumulative oil production in unconventional reservoirs.It does so with two multi-layer perceptron neural networks(MLPNN)and a least-squares support vector machine(LSSVM),hybridized with grey wolf optimization(GWO)and/or particle swarm optimization(PSO).Large,simulated datasets were divided into training(70%)and testing(30%)groups,with normalization applied to both groups.Mahalanobis distance identifies/eliminates outliers in the training subset only.A non-dominated sorting genetic algorithm(NSGA-II)combined with LSSVM selected seven influential features from the nine available input parameters:reservoir depth,porosity,permeability,thickness,bottom-hole pressure,area,CO_(2) injection rate,residual oil saturation to gas flooding,and residual oil saturation to water flooding.Predictive models were developed and tested,with performance evaluated with an overfitting index(OFI),scoring analysis,and partial dependence plots(PDP),during training and independent testing to enhance model focus and effectiveness.The LSSVM-GWO model generated the lowest root mean square error(RMSE)values(0.4052 MMT for CO_(2) storage and 9.7392 MMbbl for cumulative oil production)in the training group.That trained model also exhibited excellent generalization and minimal overfitting when applied to the testing group(RMSE of 0.6224 MMT for CO_(2) storage and 12.5143 MMbbl for cumulative oil production).PDP analysis revealed that the input features“area”and“porosity”had the most influence on the LSSVM-GWO model's pre-diction performance.This paper presents a new hybrid modeling approach that achieves accurate forecasting of CO_(2) subsurface storage and cumulative oil production.It also establishes a new standard for such forecasting,which can lead to the development of more effective and sustainable solutions for oil recovery.展开更多
Drilling operations depend on precisely controlling drilling fluid filtration volume(FV),which affects formation integrity,costs,and borehole stability.Maintaining optimal FV is essential to prevent well control issue...Drilling operations depend on precisely controlling drilling fluid filtration volume(FV),which affects formation integrity,costs,and borehole stability.Maintaining optimal FV is essential to prevent well control issues,yet forecasting it is challenging due to process complexity and measurement limitations.This study adapts machine and deep learning(ML/DL)models to predict FV in almost real-time based on more easily measured fluid properties.Radial-basis-function neural network(RBFNN),generalized regression neural network(GRNN),multilayer perceptron(MLP),convolutional neural network(CNN),and Gaussian process regression(GPR)ML models are applied to 1186 records of density,viscosity,and solids content in water-based drilling fluids deployed in fourteen wellbores.CNN outperformed other models with the lowest root mean square error(RMSE)of 0.5381 mL and demonstrated resilience to overfitting and noisy data,unlike RBFNN and GRNN.The proposed method provides reliable near-real-time FV predictions,which could be beneficial in optimizing drilling operations by helping prevent potential drilling-fluid-related issues.Fast and accurate FV forecasting from routine fluid properties represents a crucial advancement for drilling operations,highlighting the need for future dataset expansion to encompass a wider range of conditions and fluid types.展开更多
Permeability is a measure of fluid transmissibility in the rock and is a crucial concept in the evaluation of formations and the production of hydrocarbon from the reservoirs.Various techniques such as intelligent met...Permeability is a measure of fluid transmissibility in the rock and is a crucial concept in the evaluation of formations and the production of hydrocarbon from the reservoirs.Various techniques such as intelligent methods have been introduced to estimate the permeability from other petrophysical features.The efficiency and convergence issues associated with artificial neural networks have motivated researchers to use hybrid techniques for the optimization of the networks,where the artificial neural network is combined with heuristic algorithms.This research combines social ski-driver(SSD)algorithm with the multilayer perception(MLP)neural network and presents a new hybrid algorithm to predict the value of rock permeability.The performance of this novel technique is compared with two previously used hybrid methods(genetic algorithm-MLP and particle swarm optimization-MLP)to examine the effectiveness of these hybrid methods in predicting the permeability of the rock.The results indicate that the hybrid models can predict rock permeability with excellent accuracy.MLP-SSD method yields the highest coefficient of determination(0.9928)among all other methods in predicting the permeability values of the test data set,followed by MLP-PSO and MLP-GA,respectively.However,the MLP-GA converged faster than the other two methods and is computationally less expensive.展开更多
A major cause of some of serious issues encountered in a drilling project,including wellbore instability,formation damage,and drilling string stuck e which are known to increase non-productive time(NPT)and hence the d...A major cause of some of serious issues encountered in a drilling project,including wellbore instability,formation damage,and drilling string stuck e which are known to increase non-productive time(NPT)and hence the drilling cost e is what we know as mud loss.The mud loss can be prevented or at least significantly reduced by taking proper measures beforehand provided the position and intensity of such loss can be properly predicted using an accurate predictor model.Accordingly,in this study,we used the convolutional neural network(CNN)and hybridized forms of multilayer extreme learning machine(MELM)and least square support vector machine(LSSVM)with the Cuckoo optimization algorithm(COA),particle swarm optimization(PSO),and genetic algorithm(GA)for modeling the mud loss rate based on drilling data,mud properties,and geological information of 305 drilling wells penetrating the Marun Oilfield.For this purpose,we began by a pre-processing step to attenuate the effect of noise using the Savitzky-Golay method.The whole set of available data was divided into the modeling(including 2300 data points)and the validation(including 483 data points)subsets.Next,the second generation of the non-dominated sorting genetic algorithm(NSGA-Ⅱ)was applied to the modeling data to identify the most significant features for estimating the mud loss.The results showed that the prediction accuracy increased with the number of selected features,but the increase became negligible when the number of selected features exceeded 9.Accordingly,the following 9 features were selected as input to the intelligent algorithms(IAs):pump pressure,mud weight,fracture pressure,pore pressure,depth,gel 10 min/gel 10 s,fan 600/fan 300,flowrate,and formation type.Application of the hybrid algorithms and simple forms of LSSVM and CNN to the training data(80%of the modeling data,i.e.1840 data points)showed that all of the models tend to underestimate the mud loss at higher mud loss rates,although the CNN exhibited lower underestimation levels.Error analysis on different models showed that the CNN provided for a significantly higher degree of accuracy,as compared to other models.The more accurate outputs of the hybrid LSSVM model than those of the simple LSSVM indicated the large potentials of metaheuristic algorithms for achieving optimal solutions.The lower error levels obtained with the CNN model in the testing phase highlighted the excellent generalizability of this model for unseen data.The more accurate predictions obtained with this model,rather than the other models,in the validation phase further proved this latter finding.Therefore,application of this method to other wells in the same field is highly recommended.展开更多
文摘To achieve carbon dioxide(CO_(2))storage through enhanced oil recovery,accurate forecasting of CO_(2) subsurface storage and cumulative oil production is essential.This study develops hybrid predictive models for the determination of CO_(2) storage mass and cumulative oil production in unconventional reservoirs.It does so with two multi-layer perceptron neural networks(MLPNN)and a least-squares support vector machine(LSSVM),hybridized with grey wolf optimization(GWO)and/or particle swarm optimization(PSO).Large,simulated datasets were divided into training(70%)and testing(30%)groups,with normalization applied to both groups.Mahalanobis distance identifies/eliminates outliers in the training subset only.A non-dominated sorting genetic algorithm(NSGA-II)combined with LSSVM selected seven influential features from the nine available input parameters:reservoir depth,porosity,permeability,thickness,bottom-hole pressure,area,CO_(2) injection rate,residual oil saturation to gas flooding,and residual oil saturation to water flooding.Predictive models were developed and tested,with performance evaluated with an overfitting index(OFI),scoring analysis,and partial dependence plots(PDP),during training and independent testing to enhance model focus and effectiveness.The LSSVM-GWO model generated the lowest root mean square error(RMSE)values(0.4052 MMT for CO_(2) storage and 9.7392 MMbbl for cumulative oil production)in the training group.That trained model also exhibited excellent generalization and minimal overfitting when applied to the testing group(RMSE of 0.6224 MMT for CO_(2) storage and 12.5143 MMbbl for cumulative oil production).PDP analysis revealed that the input features“area”and“porosity”had the most influence on the LSSVM-GWO model's pre-diction performance.This paper presents a new hybrid modeling approach that achieves accurate forecasting of CO_(2) subsurface storage and cumulative oil production.It also establishes a new standard for such forecasting,which can lead to the development of more effective and sustainable solutions for oil recovery.
文摘Drilling operations depend on precisely controlling drilling fluid filtration volume(FV),which affects formation integrity,costs,and borehole stability.Maintaining optimal FV is essential to prevent well control issues,yet forecasting it is challenging due to process complexity and measurement limitations.This study adapts machine and deep learning(ML/DL)models to predict FV in almost real-time based on more easily measured fluid properties.Radial-basis-function neural network(RBFNN),generalized regression neural network(GRNN),multilayer perceptron(MLP),convolutional neural network(CNN),and Gaussian process regression(GPR)ML models are applied to 1186 records of density,viscosity,and solids content in water-based drilling fluids deployed in fourteen wellbores.CNN outperformed other models with the lowest root mean square error(RMSE)of 0.5381 mL and demonstrated resilience to overfitting and noisy data,unlike RBFNN and GRNN.The proposed method provides reliable near-real-time FV predictions,which could be beneficial in optimizing drilling operations by helping prevent potential drilling-fluid-related issues.Fast and accurate FV forecasting from routine fluid properties represents a crucial advancement for drilling operations,highlighting the need for future dataset expansion to encompass a wider range of conditions and fluid types.
文摘Permeability is a measure of fluid transmissibility in the rock and is a crucial concept in the evaluation of formations and the production of hydrocarbon from the reservoirs.Various techniques such as intelligent methods have been introduced to estimate the permeability from other petrophysical features.The efficiency and convergence issues associated with artificial neural networks have motivated researchers to use hybrid techniques for the optimization of the networks,where the artificial neural network is combined with heuristic algorithms.This research combines social ski-driver(SSD)algorithm with the multilayer perception(MLP)neural network and presents a new hybrid algorithm to predict the value of rock permeability.The performance of this novel technique is compared with two previously used hybrid methods(genetic algorithm-MLP and particle swarm optimization-MLP)to examine the effectiveness of these hybrid methods in predicting the permeability of the rock.The results indicate that the hybrid models can predict rock permeability with excellent accuracy.MLP-SSD method yields the highest coefficient of determination(0.9928)among all other methods in predicting the permeability values of the test data set,followed by MLP-PSO and MLP-GA,respectively.However,the MLP-GA converged faster than the other two methods and is computationally less expensive.
文摘A major cause of some of serious issues encountered in a drilling project,including wellbore instability,formation damage,and drilling string stuck e which are known to increase non-productive time(NPT)and hence the drilling cost e is what we know as mud loss.The mud loss can be prevented or at least significantly reduced by taking proper measures beforehand provided the position and intensity of such loss can be properly predicted using an accurate predictor model.Accordingly,in this study,we used the convolutional neural network(CNN)and hybridized forms of multilayer extreme learning machine(MELM)and least square support vector machine(LSSVM)with the Cuckoo optimization algorithm(COA),particle swarm optimization(PSO),and genetic algorithm(GA)for modeling the mud loss rate based on drilling data,mud properties,and geological information of 305 drilling wells penetrating the Marun Oilfield.For this purpose,we began by a pre-processing step to attenuate the effect of noise using the Savitzky-Golay method.The whole set of available data was divided into the modeling(including 2300 data points)and the validation(including 483 data points)subsets.Next,the second generation of the non-dominated sorting genetic algorithm(NSGA-Ⅱ)was applied to the modeling data to identify the most significant features for estimating the mud loss.The results showed that the prediction accuracy increased with the number of selected features,but the increase became negligible when the number of selected features exceeded 9.Accordingly,the following 9 features were selected as input to the intelligent algorithms(IAs):pump pressure,mud weight,fracture pressure,pore pressure,depth,gel 10 min/gel 10 s,fan 600/fan 300,flowrate,and formation type.Application of the hybrid algorithms and simple forms of LSSVM and CNN to the training data(80%of the modeling data,i.e.1840 data points)showed that all of the models tend to underestimate the mud loss at higher mud loss rates,although the CNN exhibited lower underestimation levels.Error analysis on different models showed that the CNN provided for a significantly higher degree of accuracy,as compared to other models.The more accurate outputs of the hybrid LSSVM model than those of the simple LSSVM indicated the large potentials of metaheuristic algorithms for achieving optimal solutions.The lower error levels obtained with the CNN model in the testing phase highlighted the excellent generalizability of this model for unseen data.The more accurate predictions obtained with this model,rather than the other models,in the validation phase further proved this latter finding.Therefore,application of this method to other wells in the same field is highly recommended.
