Simulation of CO2-oil minimum miscibility pressure (MMP) for CO2 enhanced oil recovery (EOR) using neural networks

摘要

: CO2-oil minimum miscibility pressure (MMP) is a key parameter in CO2 enhanced oil recovery (CO2-EOR) process. This work developed a fast and vigorous mathematical method using artificial neural network (ANN) model based on genetic algorithm to predict the CO2-oil MMP which was affected by several factors (i.e. reservoir temperature, the composition of reservoir oil, and the composition of injected gas). The study evaluated the performance of the newly developed ANN-based model by the errors between the predicted values and the target values. It was found that the developed ANN model provided a reliable theoretical basis for CO2 flooding, as well as offered a guidance to the successful implementation of CO2-EOR process.