Uncertainties in Static Reservoir Models
Beucher, Helene1, Didier Renard1, Brigitte Doligez2, Marco Pontiggia3, Giuseppe Bellentani3 (1) Ecole des mines de Paris, Fontainebleau, France (2) Institut Français du Pétrole, Paris (3) Eni, San Donato Milanese (Milano), Italy
The importance of uncertainties on the prediction of the
recovered volumes and the fluid flow performance is well
known. These
uncertainties mainly come from the elaboration of the static reservoir model
which results from the combination of the following steps.
The first task consists in building the architecture through the
delineation of units. It must deal with the sparseness of the data control (few
wells possibly deviated), the complexity of the stratigraphic
description (number of homogeneous layers, presence of faults) and account for
the seismic horizons available. Each layer is then populated with lithofacies reflecting the geological environment. This
environment is characterized by the depositional sequence
as
well
as the
number of lithotypes (facies
with similar properties), their proportions, trends and arrangement.
Additional constraints on lithotype proportions are
given by interpreted seismic attributes. Each lithotype
is finally assigned some petrophysical properties.
These variables are derived from various measurements types (core porosity,
log
porosity) and must be processed differently (averaging porosity while upscaling permeability). The hydrocarbon volumes are
finally calculated above the oil water contact which governs the saturation
law.
The purpose of
this paper is to evaluate these sources of uncertainty through the study of an
actual reservoir, performed in the geostatistical
framework. These uncertainties are reproduced within a realistic layering model
(multivariate method), using relevant stochastic techniques (pluri-gaussian, object based simulations) and choosing the
adequate parameters (variogram, distribution, object
intensity). The quantification is performed on global volumes as well
as on
local criteria (maps, connectivity index).