چكيده به لاتين
Due to the limited availability of light oil reservoirs, consideration of various methods to increase the recovery from heavy oil reservoirs has been considered. One of the important challenges that can arise from these reservoirs is the high viscosity of the oil, the poor mobility of the fluid that causes the phenomenon of viscous fingiring in the flooding process. On the other hand, the high inter-level elongation causes a large part of the oil to remain inside the reservoir after water flooding. One of the methods is to inject a polymer solution into the reservoir to improve the efficiency of oil. This increases the viscosity of the water and increases the process efficiency. Adding this solution, helps to change the wettability and reduce the interfacial tension between water and oil in the reservoir, thus helping to increase the oil recovery.
In this study, experimental results for injection tests of a synthesized copolymer, consisting of acrylamide (AM) and 2-acrylamido-2-methylpropane sulfonic acid (AMPS) were applied for simulation of polymer flooding into glass micro-model. Four different flooding experiments were implemented. These included water injection, polymer injection with and without salinity, and polymer injection with 1wt% of SiO2 nano-particles. This process was simulated using COMSOL Multiphysics® modeling software. Navier–Stokes and phase field equations were solved simultaneously using finite element approach. Effects of salinity, viscosity, and nano-particles were studied on displacement efficiency of oil sample in the heterogeneous porous media. Oil recovery factor and breakthrough time were obtained and compared to those of experimental outputs. Results indicated that there exists a proper accommodation between the results. Errors ranged from 3% to 7% for oil recovery factor and 0.5% to 0.9% for the breakthrough time in various injection scenarios. Using predefined structure of the under test micromodel, star-like meshes were developed in this software for the first time. This unstructured mesh is more likely to the pore structures of the sandstone reservoirs. To increase the validity of the results, a boundary layer was designated in the fluid flow path and mesh generation was performed manually in a complicated pore space instead of automatic mesh generation. Experimental and modeling results for the residual oil saturation as a function of dimensionless time were compared. Mesh dependency studies were performed to obtain the best number of mesh elements in which the most accurate results were obtained in the minimum required time. Finally, sensitivity analysis studies were performed as a function of injection rate, mobility ratio, and viscosity and the operating conditions were analyzed in terms of capillary number. All the polymer injection scenarios led to reasonable range for this parameter which verifies the accuracy of the selected operating parameters and the modeling approach.
