This study compared the predictive performance and processing speed of an artificial neural network(ANN)and a hybrid of a numerical reservoir simulation(NRS)and artificial neural network(NRS-ANN)models in estimating t...This study compared the predictive performance and processing speed of an artificial neural network(ANN)and a hybrid of a numerical reservoir simulation(NRS)and artificial neural network(NRS-ANN)models in estimating the oil production rate of the ZH86 reservoir block under waterflood recovery.The historical input variables:reservoir pressure,reservoir pore volume containing hydrocarbons,reservoir pore volume containing water and reservoir water injection rate used as inputs for ANN models.To create the NRS-ANN hybrid models,314 data sets extracted from the NRS model,which included reservoir pressure,reservoir pore volume containing hy-drocarbons,reservoir pore volume containing water and reservoir water injection rate were used.The output of the models was the historical oil production rate(HOPR in m^(3) per day)recorded from the ZH86 reservoir block.Models were developed using MATLAB R2021a and trained with 25 models in three replicate conditions(2,4 and 6),each at 1000 epochs.A comparative analysis indicated that,for all 25 models,the ANN outperformed the NRS-ANN in terms of processing speed and prediction performance.ANN models achieved an average of R^(2) and MAE of 0.8433 and 8.0964 m^(3)/day values,respectively,while NRS-ANN hybrid models achieved an average of R^(2) and MAE of 0.7828 and 8.2484 m^(3)/day values,respectively.In addition,ANN models achieved a processing speed of 49 epochs/sec,32 epochs/sec,and 24 epochs/sec after 2,4,and 6 replicates,respectively.Whereas the NRS-ANN hybrid models achieved lower average processing speeds of 45 epochs/sec,23 epochs/sec and 20 epochs/sec.In addition,the ANN optimal model outperforms the NRS-ANN model in terms of both processing speed and accuracy.The ANN optimal model achieved a speed of 336.44 epochs/sec,compared to the NRS-ANN hybrid optimal model,which achieved a speed of 52.16 epochs/sec.The ANN optimal model achieved lower RMSE and MAE values of 7.9291 m^(3)/day and 5.3855 m^(3)/day in the validation dataset compared with the hybrid ANS optimal model,which achieved 13.6821 m^(3)/day and 9.2047 m^(3)/day,respectively.The study also showed that the ANN optimal model consistently achieved higher R^(2) values:0.9472,0.9284 and 0.9316 in the training,test and validation data sets.Whereas the NRS-ANN hybrid optimal yielded lower R^(2) values of 0.8030,0.8622 and 0.7776 for the training,testing and validation datasets.The study showed that ANN models are a more effective and reliable tool,as they balance both processing speed and accuracy in estimating the oil production rate of the ZH86 reservoir block under the waterflooding recovery method.展开更多
Subspace-based signal processing methods are fundamentally pre-trained Artificial Neural Networks(ANN)that provide the basic structure for numerous computer vision applications and explore the most promising Earth Obs...Subspace-based signal processing methods are fundamentally pre-trained Artificial Neural Networks(ANN)that provide the basic structure for numerous computer vision applications and explore the most promising Earth Observation Applications(EOA).This paper examines the fundamentals of subspacebased methods and explores the most promising algorithm for forecasting ionospheric signal delays,which was designed explicitly regarding signal and noise subspaces.The learning efficiency derived from the subspace-based components of Singular Spectrum Analysis(SSA)significantly influences the implementation of Linear Recurrent Formula(LRF)and ANN models.The proposed study introduces a novel enhancement to LRF and ANN methodologies for Global Positioning System(GPS)-Total Electron Content(TEC)forecasts based on SSA.The GPS-derived TEC at Bangalore(13.02°N and 77.57°E)location grid during sunspot cycle 25(2020)is considered for analysis.The SSA-LRF-ANN model demonstrates superior accuracy compared with the SSA-LRF,Autoregressive Moving Average(ARMA),and Holt-Winter(HW)models,achieving a correlation of 0.99,a Mean Absolute Error(MAE)of 0.55 TECU,a Mean Absolute Percentage Error(MAPE)of 7.06%,and a Root Mean Square Error(RMSE)of 0.75 TECU.Furthermore,the results and discussions section presents numerical illustrations that showcase the practical implementation of the SSA-LRF-ANN and its application.展开更多
The simultaneous increase in development in Pesawaran Regency is closely correlated with the intense competi-tion for land use.However,low policy implementation effectiveness has led to construction beyond designated ...The simultaneous increase in development in Pesawaran Regency is closely correlated with the intense competi-tion for land use.However,low policy implementation effectiveness has led to construction beyond designated spatial plan.The study used a quantitative survey using Landsat images in 2016,2019,and 2022.The data analysis techniques used geographic information systems integrated with Artificial Neural Network(ANN)and Cellular Automata(CA)models.This study aims to predict land-use change in 2031,evaluate its alignment with spatial planning,and provide guidance for controlling land-use change.The results showed that there has been an increase in land use.In 2019,built-up land reached 7,069.65 Ha.The model shows its ability to predict land simulation and transformation,where it is predicted that built-up land in 2031 will experience an increase of up to 40.10%,so development and change cannot be avoided every year.This study also suggests that decision-makers and local governments should reconsider spatial planning strategies.This study shows that there have been many land use changes from 2016 to 2022.The model shows its ability to predict simulation and land transformation.When using the model,there are many changes in the land use area in 2031.This is due to wet agricultural land turning into built-up land by almost 70%.This study shows that road network influence land-use change.The cellular automata model managed to capture the complexity with simple rules.Predictions for future research should focus on conserving wetlands and primary forests.展开更多
基金National Natural Science Foundation of China grants no.41972326 and 51774258.
文摘This study compared the predictive performance and processing speed of an artificial neural network(ANN)and a hybrid of a numerical reservoir simulation(NRS)and artificial neural network(NRS-ANN)models in estimating the oil production rate of the ZH86 reservoir block under waterflood recovery.The historical input variables:reservoir pressure,reservoir pore volume containing hydrocarbons,reservoir pore volume containing water and reservoir water injection rate used as inputs for ANN models.To create the NRS-ANN hybrid models,314 data sets extracted from the NRS model,which included reservoir pressure,reservoir pore volume containing hy-drocarbons,reservoir pore volume containing water and reservoir water injection rate were used.The output of the models was the historical oil production rate(HOPR in m^(3) per day)recorded from the ZH86 reservoir block.Models were developed using MATLAB R2021a and trained with 25 models in three replicate conditions(2,4 and 6),each at 1000 epochs.A comparative analysis indicated that,for all 25 models,the ANN outperformed the NRS-ANN in terms of processing speed and prediction performance.ANN models achieved an average of R^(2) and MAE of 0.8433 and 8.0964 m^(3)/day values,respectively,while NRS-ANN hybrid models achieved an average of R^(2) and MAE of 0.7828 and 8.2484 m^(3)/day values,respectively.In addition,ANN models achieved a processing speed of 49 epochs/sec,32 epochs/sec,and 24 epochs/sec after 2,4,and 6 replicates,respectively.Whereas the NRS-ANN hybrid models achieved lower average processing speeds of 45 epochs/sec,23 epochs/sec and 20 epochs/sec.In addition,the ANN optimal model outperforms the NRS-ANN model in terms of both processing speed and accuracy.The ANN optimal model achieved a speed of 336.44 epochs/sec,compared to the NRS-ANN hybrid optimal model,which achieved a speed of 52.16 epochs/sec.The ANN optimal model achieved lower RMSE and MAE values of 7.9291 m^(3)/day and 5.3855 m^(3)/day in the validation dataset compared with the hybrid ANS optimal model,which achieved 13.6821 m^(3)/day and 9.2047 m^(3)/day,respectively.The study also showed that the ANN optimal model consistently achieved higher R^(2) values:0.9472,0.9284 and 0.9316 in the training,test and validation data sets.Whereas the NRS-ANN hybrid optimal yielded lower R^(2) values of 0.8030,0.8622 and 0.7776 for the training,testing and validation datasets.The study showed that ANN models are a more effective and reliable tool,as they balance both processing speed and accuracy in estimating the oil production rate of the ZH86 reservoir block under the waterflooding recovery method.
文摘Subspace-based signal processing methods are fundamentally pre-trained Artificial Neural Networks(ANN)that provide the basic structure for numerous computer vision applications and explore the most promising Earth Observation Applications(EOA).This paper examines the fundamentals of subspacebased methods and explores the most promising algorithm for forecasting ionospheric signal delays,which was designed explicitly regarding signal and noise subspaces.The learning efficiency derived from the subspace-based components of Singular Spectrum Analysis(SSA)significantly influences the implementation of Linear Recurrent Formula(LRF)and ANN models.The proposed study introduces a novel enhancement to LRF and ANN methodologies for Global Positioning System(GPS)-Total Electron Content(TEC)forecasts based on SSA.The GPS-derived TEC at Bangalore(13.02°N and 77.57°E)location grid during sunspot cycle 25(2020)is considered for analysis.The SSA-LRF-ANN model demonstrates superior accuracy compared with the SSA-LRF,Autoregressive Moving Average(ARMA),and Holt-Winter(HW)models,achieving a correlation of 0.99,a Mean Absolute Error(MAE)of 0.55 TECU,a Mean Absolute Percentage Error(MAPE)of 7.06%,and a Root Mean Square Error(RMSE)of 0.75 TECU.Furthermore,the results and discussions section presents numerical illustrations that showcase the practical implementation of the SSA-LRF-ANN and its application.
基金supported by the Ministry of Education,Culture,Research,and Technology Directorate General of Higher Education,Research,and Technology grant number[2147/UN2621/PN/2022].
文摘The simultaneous increase in development in Pesawaran Regency is closely correlated with the intense competi-tion for land use.However,low policy implementation effectiveness has led to construction beyond designated spatial plan.The study used a quantitative survey using Landsat images in 2016,2019,and 2022.The data analysis techniques used geographic information systems integrated with Artificial Neural Network(ANN)and Cellular Automata(CA)models.This study aims to predict land-use change in 2031,evaluate its alignment with spatial planning,and provide guidance for controlling land-use change.The results showed that there has been an increase in land use.In 2019,built-up land reached 7,069.65 Ha.The model shows its ability to predict land simulation and transformation,where it is predicted that built-up land in 2031 will experience an increase of up to 40.10%,so development and change cannot be avoided every year.This study also suggests that decision-makers and local governments should reconsider spatial planning strategies.This study shows that there have been many land use changes from 2016 to 2022.The model shows its ability to predict simulation and land transformation.When using the model,there are many changes in the land use area in 2031.This is due to wet agricultural land turning into built-up land by almost 70%.This study shows that road network influence land-use change.The cellular automata model managed to capture the complexity with simple rules.Predictions for future research should focus on conserving wetlands and primary forests.
