Next Generation Geological Modeling for Hydrocarbon Reservoir Characterization - Advances in Data, Methods, Models and Their Applications in Geoscience - page 232

Fig. 12. Validation of QuantUM AHM workflow with dynamic inversion of Brugge
synthetic model (Peters et al. 2009): a) and b) algorithm convergence diagnostics, (negative)
entropy and objective function, respectively, c) and d) well water-cut response curves,
obtained with an ensemble of prior and posterior models, respectively, e) and f) ensemble-
averaged statistical estimators, mean and variance, respectively and g) and h) prior and
posterior log-permeability distribution maps of model top layer, respectively
are conditioned to the data as well as approximate the forecast uncertainty. But the crux of
the matter here is two-fold: a) throughout the inversion, some model realizations may have
created non-geologically realistic features and b) many of the underlying geological
parameters may have an insignificant effect on recovery performance.
In addition to the traditional, single-parameter sensitivity studies to identify the important
and geologically relevant parameters, a more sophisticated version uses streamline