Road to Bio-polymer Flooding in Carbonate Reservoirs: Numerical Modelling

Polymer flooding can be of pivotal importance in improving oil recovery factors from carbonate reservoirs. Bio-based polymer floods have displayed excellent and cost-effective enhanced oil recovery (EOR) performance at laboratory-scale experiments. Nevertheless, despite these promising performance signs, many bio-based-polymer flooding options fail to progress outside the experimental stages. This paper’s thrust is to numerically assess the performance of an okra-based solution for polymer flooding across different scales. The numerical models aim to reproduce the results from laboratory core-flooding experiments that were previously published. Furthermore, this study conducts a sensitivity assessment examining the factors that significantly impact the flood’s outcome. The models are also used to explore oil displacement and the propagation of the okra-based polymer. The numerical model was built on CMG STARS using a one-dimensional (1D) approach to match lab-scale flooding experiments. The laboratory examination was designed to be applied at reservoir conditions in western Kazakhstan. CMOST was used for history matching the simulation outputs with the lab-scale oil production volumes. The procedure honored the upper and lower range limits of the following parameters: polymer viscosity and concentration, water and polymer injection rates, along with a relative permeability model for both for water flooding (WF) and polymer flooding (PF) conditions. A sub-model experiment design (DoE) was generated explicitly using the Latin hypercube sampling (LHS) method. The numerical simulation results showed that the laboratory-obtained final oil recovery numbers for WF exceeded the corresponding numerically-obtained recovery. However, in terms of the overall performance, the difference between the cumulative oil production between the laboratory experiments and numerical model is only four percent. This is due to the rate alteration required under laboratory conditions. The sensitivity assessment indicated that the relative permeability curvature interpolation and the polymer concentration are the most dominant independent parameters impacting the final oil recovery (in the homogeneous model). For a 2D model, the propagation of the polymer front remains uniform compared to the more pointed water fronts in the WF. A significant reduction in the total mobility from 10 mD/cp to 5.2 mD/cp confirms the sweep efficiency enhancement. The history-matching results suggest that the saturation endpoints have been altered during PF in comparison to the WF. This behavior was reported as inconclusive for several polymer-flooding options. The meta-model’s sensitivity assessment highlights a strong relationship between the polymer concentration and the final oil recovery for the okra-based polymer flooding.