Matching Both
Production History and 4D-Seismic Attributes
Le Ravalec-Dupin, Mickaële1,
Yann Gautier2, Alexandre Fornel1, Frédéric Roggero3
(1) IFP,
One of the most challenging fields in
reservoir engineering encompasses the integration of all available data in
models for the characterization of reservoirs and the reduction of
uncertainties in oil and gas production. A reservoir model is a grid populated
by petrophysical properties such as porosities and permeabilities. During the
field life, many data are collected including production data and since the
late 1990s 4D seismic data. We present a general matching workflow to build
reservoir models constrained to both production data and 4D seismic data.
First, a high resolution 3D geological model is simulated, then upscaled.
Secondly, the flow simulation is run. On one hand, it yields simulated
production answers, which can be compared to the actual production data. On the
other hand, it produces saturation and pressure grids, which are downscaled to
the level of the fine geological model. Then, a petroelastic model is used to
derive impedances, which are filtered before being compared to the actual
impedances. The high resolution model is subsequently perturbed and the entire
workflow is repeated until production and 4D seismic data are matched. The
gradual deformation method is used to govern model perturbations. Thus, only a
limited number of parameters is required to vary the whole porosity and
permeability fields and the spatial structure is preserved. The matching
process involves the computation of gradients, which is a CPU-time consuming
step. The efficiency of the whole process is ensured by distributing these
computations on several computers. Using a 3D synthetic case, we evidence the
interest and efficiency of our workflow. Numerical tests are performed to
quantify the influence of the 4D seismic information.
AAPG Search and Discover Article #90063©2007 AAPG Annual Convention, Long Beach, California