Abstract
Condensate banking around wellbores can significantly shorten the production from gas-condensate reservoirs. Several approaches to mitigate this issue have been proposed in the literature, among which gas injection comes out with promising results. This method's efficiency has been largely investigated at core and field scales, but the underlying flow phenomena taking place at the pore-scale are still poorly understood. In order to address this gap, a compositional pore-network model was used to reproduce gas injection in a sandstone sample following condensate accumulation. C-1, C-2, CO2, N-2 and produced gas were tested as the candidates for condensate banking remediation at different pressure levels. After gas flooding, condensate saturation, heavy component recovery and gas relative permeability were quantified to appraise the achieved gas flow improvement. The results indicated that C-2 and CO2 were, overall, the most effective gases to clear the accumulated condensate and re-establish the gas flow. At high pressures, their injection reduced the liquid saturation in the porous medium at least in 80%, and restored gas relative permeabilities to values close to unity. At low pressures, however, no tested gas injection composition could reduce the condensate content in the porous medium more than 30%, contradicting the estimates based solely on liquid dropout curves. With the presented analyses, this work sets out the pioneer efforts towards an accurate pore-scale modeling of miscible gas injection in gas-condensate reservoirs.
NSTL
Elsevier