Application of the ensemble Kalman filter for history matching of reservoir simulation models

History matching (HM) is the process of adjusting the reservoir simulation model parameters so that the computed values of observables such as rates or pressures at individual wells are in reasonable agreement with actual measurements of those quantities. Improved model is used to predict the oil production and for optimizing field development.
Сonventional algorithms of HM can be reduced to the minimization of complex multimodal functions. Such optimization problem and the following problem of oil production forecast uncertainty quantification became extremely complicated. Ensemble Kalman Filter gives a transparent recipe for sequental data assimilation and simultaneous uncertainty quantification.
In this report the software implementation of this method is considered and the results of the numerical calculations are demonstrated. It is shown that the developed package allows us to successfully solve HM-problems and has advantages over more traditional ones.