Assisted History Matching for Surface Coupled Gas Reservoir Simulation

Abstract

Abstract For a gas reservoir, the intake pressures are available at a delivery point and the operator is obligated to supply gas honoring such downstream pressure requirements. As a result, it is imperative that the reservoir deliverability prediction be coupled with the pressure drop in the surface network. Additionally, this coupled system should not only honor the imposed delivery point pressure constraints but also the historical attributes of pressures and rates. In this paper, a modified Gauss-Newton method is utilized in conjunction with a non-linear parameter estimation algorithm to history match a surface-reservoir coupled gas reservoir simulation. Control Volume Finite Element (CVFE) based reservoir simulator is amenable to unstructured grids to resolve near wellbore and high activity (high permeability channels, fractures) flow areas. The pressure drop in the surface network is modeled using Weymouth's equation of steady state pipeline flow. More importantly, the total system is solved in a coupled fashion thereby enabling the solution to be controlled by surface pressure constraints. In addition, both surface and reservoir decision variables are estimated using a modified Gauss Newton algorithm in the assisted history matching step. Examples from dry gas reservoir demonstrate the usefulness of this methodology. First, the necessity of coupling surface and reservoir models is highlighted. Next, the computational efficiency of the assisted history matching algorithm as compared to perturbing critical reservoir and surface attributes in a heuristic manner is underscored. There are three significant contributions of this paper, namely (a) an assisted history matching algorithm proves efficient in comparison to arbitrary perturbation of decision variables (b) a coupled reservoir-surface model renders a more accurate pressure prediction of the total system and finally, (c) an unstructured grid simulator provides the platform for accurate and cost-effective reservoir modeling alternative.