DEVELOPMENT OF ACCURATE MODELS FOR ESTIMATION OF PURE CO₂-OIL MINIMUM MISCIBILITY PRESSURE BASED ON ARTIFICIAL INTELLIGENCE METHODS

Design of CO₂ miscible flooding for enhanced oil recovery depends on CO₂-oil Minimum Miscibility Pressure (MMP) which is determined either experimentally or by analytical methods. Some efforts are recently conducted to calculate MMP based on Artificial Intelligence (AI) methods. In this chapter, an Artificial Neural Network (ANN) is developed to predict the CO2-oil MMP in the case of pure CO2 injection, and a previous Genetic Algorithm (GA) method is improved by using Particle Swarm Optimization (PSO) method. Analysis of the results shows that the ANNbased model yields a better match with the experimental data.