Development of a fully implicit approach with intensive variables for compositional reservoir simulation

Abstract

Compositional reservoir simulators are important tools for application in enhanced oil recovery processes. These simulators solve partial differential equations arising from modeling fluid flow in permeable media. Various algorithms for the solution of such partial differential equations are available, which their application greatly impacts the computational performance of the simulators. In this work, a new fully implicit approach called PZS (pressure, overall composition, and water saturation) is proposed and implemented. The new formulation considers pressure, water saturation, and overall compositions as primary variables, reducing the number of unknowns by one when compared to other volume balance fully implicit formulations. The new approach is obtained by a variable change and elimination of a well known volume balance approach. The new approach is implemented in the UTOMPRS simulator that has been developed at The University of Texas at Austin for simulation of several multicomponent/multiphase recovery processes. The PZS formulation is compared to the volume balance based and IMPEC (Implicit Pressure Explicit Composition) formulations in UTCOMPRS. We observe that the PZS approach is, in general, faster than the other approaches tested.