An upscaling workflow for recovery evaluation in heavy oil reservoirs
conference paper
The displacement of heavy oil by water flooding is characterized by instable features known as viscous fingers. These fingers may form at very small scales, as can be observed in laboratory experiments and predicted from instability theory. Explicit representation of such features in large-scale reservoir models is practically impossible because of computational limitations. Here we present a workflow for estimation of effective flow properties (in this case, relative permeability and capillary pressure) that enable evaluation of field-scale recovery by reservoir simulation on coarse grids. The workflow is applied to an example based on real data. Laboratory experiments conducted on a thin sandstone slab filled with heavy oil are simulated using a very fine 2D grid; an extension to 3D simulations is discussed in a companion paper by Hofstee et al. Flow properties are adjusted to reproduce key observed characteristics of the displacement. The simulation domain is subsequently extended in order to compute the changes in displacement characteristics over time. We present a number of useful diagnostics that could be used to provide first order predictions on field scales. The recovery efficiency curve as obtained on a fine grid is taken as the target for an upscaling step, based on an experimentally observed self-similarity in this profile
TNO Identifier
526050
Article nr.
WHOC15-373
Source title
World Heavy Oil Contgress 2015, Edmonton, Alberta, Canada
Pages
1-9
Files
To receive the publication files, please send an e-mail request to TNO Repository.