Skip to Main content Skip to Navigation
Journal articles

3D line-support grid flattening for more accurate geostatistical reservoir population with petrophysical properties.

Abstract : Reservoir models are essential if we need to clearly understand the fossil resources and, hence, to make better use of them. Feeding these models with physical properties on the basis of wells data is a key step in their construction. Line-support (LS) grid is the most popular grid in reservoir engineering, it is massively used for reservoir simulations. In the current methods used to populate with properties the LS grid of a reservoir unit, a Cartesian grid of equivalent size (in each direction), obtained by averaging the edge lengths, is first of all completed. The properties calculated in this way are then transferred as they are into the initial LS grid, because there is cell-for-cell correspondence. This leads to distortion of the Cartesian grid, making it fit the shape of the LS grid. This has the effect of altering calculations of correlation distances between well markers in geostatistical population simulations. Consequently, this primarily induces distortions on the simulated bodies. To resolve this problem, in this paper, we propose innovative methods for a "smooth" conversion from the LS grid of the structural space to the Cartesian grid of the geostatistical population space. The basic principle is to calculate the correlation distances between wells on the basis of "quasi-isometric" flattening of the stratigraphic unit LS grid in the population space. This same flattening technique is then used for inverse transfer of the properties from the population space to the structural space.
Complete list of metadata
Contributor : Françoise Bertrand Connect in order to contact the contributor
Submitted on : Monday, July 28, 2014 - 2:34:25 PM
Last modification on : Sunday, June 26, 2022 - 4:49:53 AM





Chakib Bennis, Houman Borouchaki, Cyrielle Dumont, Olivier Lerat, Mathieu Poudret, et al.. 3D line-support grid flattening for more accurate geostatistical reservoir population with petrophysical properties.. Engineering with Computers, Springer Verlag, 2014, 30 (3), pp.403-421. ⟨10.1007/s00366-012-0311-9⟩. ⟨hal-01052688⟩



Record views