Effect of Dispersion Coefficient on Polymer Injection Studied with Random Walk Particle Tracking Method

Durgut İ. , Mamak G.

Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, vol.10, no.1, pp.432-438, 2021 (National Refreed University Journal)

  • Publication Type: Article / Article
  • Volume: 10 Issue: 1
  • Publication Date: 2021
  • Title of Journal : Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi
  • Page Numbers: pp.432-438


Polymer injection is a chemical EOR process, where the aim is to improve sweep efficiency of water flooding in an oil reservoir by increasing the water viscosity with the thickening effect of polymer injected with water. A random walk particle tracking model is developed to simulate the injection of polymer into an oil reservoir and it is integrated into an open source black-oil reservoir simulator (SINTEF’s MATLAB Reservoir Simulation Toolbox, MRST). The black-oil simulator solves the set of partial differential equations describing multiphase fluid flow in the porous medium, whereas the transport of the injected polymer due to advection and dispersion processes is obtained from the applied random walk method. The finite difference/finite volume discretizations of the continuity equations applied in MRST does not involve the physical dispersion processes. The dispersion observed on the results are called the numerical dispersion and are due to the numerical discretization methods applied to solve the continuity equations. Since dispersion in porous media is a scale dependent process, it is hard to quantify its coefficient by experiments or empirical equations in general. The random walk method to model the transport of injected polymer in the reservoir is independent of discretization. Hence, it does not involve numerical dispersion. Instead, the physical dispersion due to, for instance, the tortuosity of flow path or the adsorption of polymer in porous media can be included into the model. Therefore, we used the method to analyze the effect of dispersion coefficient. Increased dispersion coefficient causes the results to converge to actual MRST solution with increased uncertainties of polymer concentration. The dispersivity of a reservoir in simple one-dimensional problems can be determined with the method.