Minimum miscibility pressure prediction using support vector machines

Miscible gas injection processes are among the effective methods for enhanced oil recovery. 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 time-consuming, searching for fast and robust mathematical determination of gas-oil MMP is usually requested. This paper introduces Support Vector Machines (SVM), a relatively new powerful machine learning method based on statistical learning theory, into MMP forecasting. The validity of this new model was successfully approved by comparing the model results to the experimental gas-oil MMP and the calculated results for the common gas-oil MMP correlations. The new model yielded the accurate prediction of the experimental gas-oil MMP with the lowest average relative and average absolute error among all tested gas-oil MMP correlations. In addition, the new model could be used for predicting the gas-oil MMP at higher fractions of non-CO2 components

Keywords:

Support vector machines (SVM) , Minimum miscibility pressure (MMP) , Enhanced oil recovery (EOR) , Miscible gas flooding