Study of Inverse Modeling to Infer Permeability Distribution of Geothermal Reservoir Based on Production Hystory Data Using Ensemble Kalman Filter

Abstract

The research is accomplished to evaluate the application of EnKF method in the reservoir geothermal simulation based on production history data. Two experimental models are constructed and simulated using TOUGH2 reservoir simulation. One of the experimental models is assigned with single permeability in all grids, while the other one is assigned with various permeability value distributed in all grids of model. It is also performed several data assimilation scenarios. This research indicates that EnKF method can be applied in history matching process to inverse model permeability from geothermal reservoir. It also shows that pressure data assimilation gives a better matching than temperature data assimilation. However, EnKF method may give non-unique outcomes or solutions for the very great range of permeability in a model. The addition of extra information (a priori information) to the inversion process or data assimilation is proven to diminish the effect of non-uniqueness.