CO2 Minimum Miscible Pressure MMP Estimation using Multiple Linear Regression MLR Technique

Abstract

Abstract Miscible gas injection processes are among the effective methods for enhanced oil recovery (EOR). A key parameter in the design of gas injection project is the minimum miscibility pressure (MMP), whereas local displacement efficiency from gas injection is highly dependent on the MMP. Because experimental determination of MMP is very expensive and timeconsuming, searching for fast and robust mathematical determination of gas-oil MMP is usually wished. A new CO2 MMP correlation based on Multiple-Linear-Regression modeling (MLR) technique has been successfully developed to more accurately estimate the CO2 MMP for a wide range of live and heavy crude oils. The newly developed CO2 MMP correlation is originated from CO2 MMP experimental data in addition to database from the worldwide published literature that covers 103 pure CO2 MMP data for various live and dead oil samples. The proposed model is trained by exploiting 80% (82 data points) of the data bank. This correlation is expressed as a function of reservoir temperature, C7+ molecular weight, mole fractions of; non-hydrocarbon components (CO2, H2S and N2), direct correlating components (C1, C2 and C5) and indirect correlating components (C3, C4, C6 and C7). Further, to investigate the authenticity in depth, the proposed correlation is compared with five most commonly used pure CO2 MMP correlations from the literature. A statistical comparison is performed for both training data set (82 data points) as well as testing data set (21 data points). It is found that the proposed CO2 MMP correlation provides the best reproduction of MMP data with a percentage average absolute error of 6.083% and 6.328% for training and testing categories respectively. Finally, the proposed model shows great performance using new set of samples compared with other correlations.