Model Dimensionality Reduction by Upscaling of the Decomposition Spectra of Its Static Parameters

Abstract

In this paper we propose a novel method of reducing the number of history matching variables required for adequate approximation of static properties associated with reservoir models of hydrocarbons. The method is based on the expansion of the static parameters into discrete orthogonal basis with subsequent parametrization of underling expansion coefficients. The method is demonstrated with application to a simple synthetic example and a sector model of the real oil field. It is shown that utilization of the method described here provides additional degree of model reduction by a factor of ten or more.