Numerical Simulation of the Impact of Natural Fracture on Fluid Composition Variation Through a Porous Medium

Abstract

In order to lessen the computational time in fractured oil reservoir simulations, all fractures are usually assumed to be as one equivalent fracture at the center or around the model. This, specially, has applications in industrial engineering software, where this assumption applies. In this study, using two general contradictory examples, it is shown that ignoring a fracture network and assuming an equivalent single-fracture has no logical justification and results in a considerable error. The effect of fracture aperture on composition distribution of a binary and a ternary mixture was also investigated. These mixtures were C1 (methane)/n-C4 (normal-butane) and C1 (methane)/C2 (ethane)/n-C4 (normal-butane), which were under diffusion and natural convection. Governing equations were numerically solved using matlab. One of the main relevant applications of this study is where permeability and temperature gradient are the key difference between reservoirs. Compositional distribution from this study could be used to estimate initial oil in place. Using this study, one can find the optimum permeability, namely the permeability at which the maximum species separation happens, and the threshold permeability (or fracture aperture), after which the convection imposes its effect on composition distribution. It is found that the threshold permeability is not constant from reservoir to reservoir. Also, one can find that full mixing happens in the model, namely heavy and light densities of top and bottom mix up together in the model. Furthermore, after maximum separation point, convection causes unification of components.