Modeling and simulation of polymer flooding including the effects of fracturing
Chemical enhanced oil recovery (EOR) technology has attracted increasing interest in recent years with declining oil production from conventional oil reserves. Water flooding of heterogeneous reservoirs with viscous oil leaves considerable amount of remaining oil even at high producing water cuts. Polymer flooding is a mature EOR technology for augmenting recovery of moderately viscous oil. Water soluble polymers are used to reduce water mobility and improve sweep efficiency. For very viscous oil, polymer flooding is a potential non-thermal approach for minimizing viscous fingering and improving both displacement sweep efficiency and volumetric sweep efficiency. Polymer manufacturing techniques has been significantly advanced since 1980’s, which provides improved polymer quality and keeps polymer price relatively low. Compared with unconventional oil recovery techniques such as hydraulic fracturing, well planned and optimized polymer flooding can be profitable even at pessimistic oil price. It is thus crucial to have a reservoir simulator that is able to accurately model polymer properties and simulate polymer flooding in complex reservoir systems.
Polymer rheological behavior is dependent on polymer molecular structure, concentration, Darcy velocity, brine salinity, hardness, permeability, porosity, etc. We improved polymer rheology modeling for heterogeneous reservoirs where permeability varies for orders of magnitude. For an injection well, a large portion of pressure drop is lost near wellbore where apparent polymer viscosity as a function of Darcy velocity varies drastically. Conventional analytical well models fail to capture the non-Newtonian effect of apparent polymer viscosity and make injectivity predictions widely deviated from true solutions especially for coarse-grid simulations. We developed a semi-analytical polymer injectivity model and implemented it into UTCHEM. This model is able to handle both shear-thinning and shear-thickening polymer rheology. It successfully avoids the grid effect and matches fine-grid simulation results and analytical solutions. Another challenge is to model polymer injectivity under fracturing conditions. To maintain an economic polymer injection rate, wellbore pressure may exceed the fracture initiation pressure. We developed a framework to couple a fracture model with UTCHEM. This coupled simulator is able to model fracture propagation during polymer injection. Finally several simulation studies were conducted to show the impacts of polymer rheological behavior, loss of polymer into aquifer, near wellbore effect and fracture propagation.