Numerical interpretation of SCAL data with automated history matching tool

The relative permeability and capillary pressure are interdependent characteristics of two phase flow which can be most accurately evaluated by performing flooding and centrifuge experiments on the same core sample. The presented approach allows to history match the results of both experiments at the same time with synchronous determination of optimal parameters for capillary pressure and relative permeability curves. The accuracy and robustness of the approach are ensured by NURBS representation of the curves and by adjoint method for calculation of sensitivity coefficients, meaning derivatives of the data which we are trying to match with respect to the model parameters. Numerical modeling takes into account interference of both viscous and capillary forces which makes the interpretation of experimental results more accurate compared to well-known analytical approaches. During the history matching process not only shape parameters of the above-mentioned curves is important to match, but also due to the uncertainty of residual oil saturation and connate water saturation, these values can be calculated more accurately by simulation as well. In this paper, in-house made tool based on publicly available MRST simulator is presented. Simulated cases are based on synthetic data and include several scenarios with various combinations of experiments.