Prediction of Oil Recovery Factor in CO2 Injection Process

In this study, prediction of recovery factor (RF) for CO₂ injection into oil reservoirs based on artificial neural networks (ANNs) and mathematical models were investigated. To design the optimum ANN model, number of neurons, hidden layers, and training function were studied. Finally, efficiency of the models was evaluated using new data. As a result of this work, it can be concluded that it is possible to predict RF in CO₂ injection process by ANN and mathematical model. However, values that obtained from ANN were in the best agreement with the actual values than regression model. The proposed artificial neural network predicted RF during CO₂ injection with error about 0.396%.