A NEW MODEL FOR ESTIMATING THE NON-DARCY FLOW COEFFICIENT USING GENETIC PROGRAMMING

Abstract

The researcher studied the development of a new model to predict the non-Darcy flow coefficient with high accuracy compared to the commonly used correlations. This is a major source of rate dependent pseudo skin around wellbore. The researcher built the new model using genetic programming. Where the input of the new model is the permeability and viscosity of the gas and the output is the non-Darcy flow coefficient. The new model was built using 450 points for the Beta Coefficient (Turbulence Coefficient) obtained from multi-rate wells tests. This data is divided into two groups. The first group, consisting of 298 points, was used to construct the new model. The second group, consisting of 152 points, was used to test the new model. The results indicate that the new model is suitable for estimating the non-Darcy flow Coefficient more accurately than other commonly used empirical correlations and to obtain more reliable inflow performance relationships.