Compositional Modeling of Retrograde Gas-Condensate Reservoirs in Multimechanistic Flow Domains

Abstract

Abstract A multi-mechanistic flow environment is the result of the combined action of a Darcian flow component (the macroscopic flow of the phase due to pressure gradients) and a Fickian-like or diffusive flow component (diffusive flow due to molecular concentration gradients) taking place in a hydrocarbon reservoir. The present work presents the framework needed for the assessment of the impact of multi-mechanistic flow on systems where complex fluid behavior—such as that of retrograde gas-condensate fluids—requires the implementation of compositional reservoir simulators. Due to the complex fluid behavior nature of gas-condensate fluids, a fully-implicit (IMPISC-type) compositional model is implemented and the model is used for the study of the isothermal depletion of naturally fractured retrograde gas reservoirs. In these systems, especially those tight systems with very low permeability (k> 0.1 md), bulk fluid flow as predicted by Darcy's law might not take place despite the presence of large pressure gradients. The use of an effective diffusion coefficient in the gas phase—which accounts for the combined effect of the different diffusion mechanisms that could take place in a porous medium—and its relative contribution to fluid recovery is discussed. The compositional tracking capabilities of the model are tested and the conditions where Fickian flow can be the major player in recovery predictions and considerably overcome the flow impairment to gas flow posed by the eventual appearance of a condensate barrier—typical of gas-condensate systems—are investigated. Finally, a mapping that defines different domains where multi-mechanistic flow can be expected in compositional simulators of retrograde gas-condensate reservoirs is presented. Introduction In typical natural gas reservoirs, all hydrocarbons exist as a single free gas phase at conditions of discovery. Depending on the composition of the initial hydrocarbon mixture in place and their depletion behavior, we recognize up to three kinds of natural gas reservoirs: dry gas reservoirs, wet gas reservoirs, and retrograde gas or gas-condensate reservoirs. The latter is the richest in terms of heavy hydrocarbons; and thus, it is very likely to develop a second heavier hydrocarbon phase (liquid condensate) upon isothermal depletion. This situation is illustrated by Figure 1. In contrast, dry gases and wet gases do not undergo phase changes upon reservoir depletion, as their phase envelope's cricondetherms are found to the left of the reservoir temperature isobar line. Craft and Hawkins[1] demonstrated how the trend to gas and retrograde gas reservoir discoveries (with respect to oil reservoir discoveries) have been increasing ever since deeper drilling started in the 1950s. Since then, retrograde gas reservoirs have grown in importance through time. The performance behavior of this class of reservoirs is usually modeled using compositional simulators because their depletion performance is highly influenced by changes in fluid composition. Often times, highly sophisticated and computationally intensive compositional simulations are needed for the accurate modeling of their performance, phase behavior and fluid flow characteristics.