Review of nonlinear kalman, ensemble and particle filtering with application to the reservoir history matching problem

This chapter reviews the recent advances in Bayesian filtering approaches, with the focus on those suitable for data assimilation in high-dimensional systems. We discuss the similarities and differences of these filtering approaches, and compare their performance in an application to the history matching problem in a synthetic, twodimensional, oil-water reservoir model.

Data Assimilation Ensemble Kalman Filter Gaussian Sum Filter History Matching in Reservoir Models Nonlinear Kalman Filter Particle Filter Sequential Bayesian Filtering