Numerical Simulation of 2D Hot Waterflooding Using Analytical Solutions Along Streamlines

Abstract

Streamline simulation is an alternative method to conventional finite difference methods, more suited to complex reservoir physics but limited in size by computational effort. The main feature of streamline modeling is replacing the space variable by the time-of-flight. In this approach, the fluid transport equation is solved along each streamline. This method is more attractive in convection-dominated flow and in large heterogeneous models. In high paraffinic oil reservoirs cold water flooding does not lead to satisfactory ultimate recovery factors due to the precipitation of solid paraffin from the liquid in the pore space. For this particular case, hot water flooding is an enhanced oil recovery technique option to increase oil production. In this work the process of displacement of waxy crude by a hot water slug followed by cold water drive in porous media was modeled. The liquid phase is composed by two pseudo-components, oil and paraffin. It was assumed that the viscosities of the components and paraffin concentration in the liquid phase are functions of temperature only. The simulations were run in 2D incompressible models without heat losses to adjacent layers and neglecting gravity and capillary effects. The system of hyperbolic equations that represents the mass conservation of each component was solved analytically. The solution of the continuous injection problem is self-similar, but in the case of slug injection interaction between waves of different families occurs. An important characteristic arising from the use of analytical solutions along streamlines is the ability to capture correctly the saturation and temperature shocks. Using analytical solutions along streamlines allowed to reduce the number of time steps and achieved results free of numerical diffusion. The results of streamline simulations were compared a commercial finite difference based simulator and presented close agreement with lower computational cost. The simulations were performed for two cases. The first one considers the injection of a high temperature water slug followed by water drive at the reservoir initial temperature. The second case analyzes the injection of a water slug at a lower temperature followed by water drive at a temperature higher than the reservoir temperature.