Solution Error Models: A New Approach for Coarse Grid History Matching

摘要

This paper presents a new technique – solution error models –and applies it to the problem of history matching using simplecoarse grid models. The solution error model technique isbased on building a statistical model that describes the errorsintroduced in the simplified solution. The difference betweenobservations and the coarse grid solution is measured by the“goodness of fit” (or misfit) term. The results show that astandard least squares approach to quantifying the misfitfunction yields significantly biased estimates of parameters, ifthe underlying model is biased. The bias can be almostcompletely corrected through use of the solution error model.The solution error model also introduces the use of timedependent covariance, instead of the usual single valuedvariance. This improves confidence in estimates of reservoirparameters.