Abstract
The accurate and efficient large-scale simulation of chemical flooding using mechanistic reservoir simulation models has been a big challenge for the past several decades, but recent advances in both the modelling of physical properties and numerical algorithms have finally made it possible to use mechanistic chemical simulators for field-scale application. In this paper, a new version of UTCHEM built in an adaptive implicit framework was used to simulate chemical floods in large heterogeneous oil reservoirs. The new developments made it feasible to do simulations using several million gridblocks with sufficiently small gridblock sizes to ensure physical accuracy. Four case studies of polymer flooding and surfactant-polymer flooding are presented. The grid sizes range from 800,000 to 3.7 million gridblocks. The incremental enhanced oil recovery is illustrated by comparison with waterflooding. The polymer model takes into account the effects of shear thinning, adsorption, and salinity among other phenomena. The surfactant model takes into account all Winsor microemulsion phase behavior types, salinity effects, and capillary desaturation among other process phenomena. The surfactant-polymer floods are designed with a salinity gradient.
