Reservoir characterization with pore network extraction applied to presalt analog rocks

Daniel Plucenio¹; Leonardo Verbicaro²; Webster Mohriak³; Jaqueline Saad²; Tadeu Manoel⁴; Leonardo Rangel⁴

¹ Universidade Federal de Santa Catarina & ESSS; ² Concremat Engenharia; ³ Universidade do Rio de Janeiro & Concremat Engenharia; ⁴ ESSS

doi:10.20906/CPS/CILAMCE2015-0745

Resumo

Methods based on image analysis are proving to be a powerful ally for reservoir rock characterization with advantages such as analysis uniformity and repeatability as well as reduction of operational costs and time. Once a three-dimensional image is obtained, petrophysical properties can be estimated using a mathematical model that captures the physics of fluid flow through porous media. This work uses the Pore network extraction methodology, which in essence consists in creating a topologically equivalent structure of the porous system but with a simplified geometry, trying to mimic the rock in its nature, which contains larger spaces between grains (pores) and elongated apertures connecting two or more pores (throats). Fluid flow is simulated by treating the network as a hydraulic system. Shape factors are used to capture the flow resistance of the original porous structures. A system of equations is composed using generalized Poiseulle's equation in order to find pressure values for each pore. The main advantage of this methodology is the capability of analyzing larger images, which are becoming increasingly necessary in the characterization of carbonate rocks, whose computational costs required by other existing methods are very high and often prohibitive. We present the application of some existing techniques such as euclidean distance transformation and medial axis extraction to provide improvements in some of the stages of the network generation. Pore networks generated from siliciclastic rock images are shown as well as the results obtained with the model for absolute permeability, porosity and capillary pressure curve with mercury injection which are compared with experimental data with good agreement. Networks and results generated from carbonate rocks similar to those found in Pre-salt reservoirs are presented and we discuss the challenges of carbonate characterization with image analysis.

Palavras-chave: Pore Network; Image analysys; Digital Rock; Rock Characterization; Medial Axis extraction