An EOS-Based Compositional Thermal Reservoir Simulator

Abstract

Abstract In the past thirty years, the development of compositional reservoir simulators using various equations of state (EOS) has been addressed in the SPE literature. However, the development of compositional thermal simulators in conjunction with EOS formulation has been ignored, in particular. Therefore, a fully implicit, parallel, compositional EOS-based simulator has been developed. In this model, an equation of state is used for equilibrium calculations among all phases (oil, gas, and aqueous). Also, the physical properties are calculated based on an EOS, hence obviating the need for using steam tables for calculation of water/steam properties. The governing equations for the model comprise fugacity equations between the three phases, material balance, pore volume constraint and energy equations. The governing partial differential equations are solved using finite difference or finite volume approximations. In the steam injection process, the solubility of oil in water-rich phase and the solubility of water in oil phase can be high. This model takes into account the solubility of water in oil phase and the solubility of hydrocarbon components in water-rich phase, using three-phase flash calculations. This simulator can be used in various thermal flooding processes (i.e. hot water or steam injections). Since the simulator was implemented for parallel computers, it is capable of solving large-scale thermal flooding problems. The simulator is successfully validated using analytical solutions. Also, simulations are carried out to compare this model with commercial simulators. The use of an EOS for calculation of various properties for each phase automatically satisfies the thermodynamic consistency requirements. On the other hand, using the K-value approach, which is not thermodynamically robust, may lead to results that are thermodynamically inconsistent. This simulator accurately tracks all components and mass transfer between phases using an EOS; hence, it will produce thermodynamically consistent results and project accurate prediction of thermal recovery processes. Introduction This paper describes the development of a fully implicit EOS thermal flooding simulator. The objective of this work is to model the thermodynamical properties and compositional effects on phase behavior calculation more accurately. In this model, all phases (oil, gas, and aqueous) are in equilibrium, and the equation of state is used for determining equilibrium between the phases. Physical properties of the phases and the components are calculated with the EOS. It is not necessary to use any steam-table to find water/steam properties. The difference between the K-value approach and the EOS-based thermal models lies in the fact that the K-value is not a function of composition of the phases. The models, which use the K-value method, do not consider the solubility of hydrocarbon components in the aqueous phase and the solubility of water in the hydrocarbon liquid phase. However, as the temperature increases, the solubility of hydrocarbon components in the aqueous phase and also the solubility of water in the hydrocarbon liquid phase are significantly increased.