Stochastic Pore Network Generation from 3D Rock Images

• DOI: 10.1007/s11242-011-9792-z
• Publication Date: 2011-06-23T22:47:10Z
• Authors: Z. Jiang; M. I. J. van Dijke; K. Wu; G. D. Couples; K. S. Sorbie; J. Ma
• Keywords: Relative permeability; Network model; Capillary pressure

Pore networks can be extracted from 3D rock images to accurately predict multi-phase flow properties of rocks by network simulation. However, the predicted may sensitive pore if it is small, even though its underlying characteristics are representative. Therefore, a challenge investigate effects on microscopic features individually and collectively based small samples. In this article, new approach introduced generate an initial stochastic arbitrary size that has same as parent network. Firstly, we characterise realistic in terms distributions geometrical correlations between these properties, well connectivity function describing detailed topology. Secondly, create size, required number nodes bonds with correlated original The randomly located given domain connected according strongest correlation node bond while honouring function. Thirdly, using state-of-the-art two-phase model, demonstrate for two samples (capillary pressure, absolute relative permeability) preserved networks, particular, latter larger than original, or method reveals sample not We also show information necessary reproduce correctly, particular This forms basis generation multiple at different resolutions combining relevant statistics corresponding which will presented future publication.