45799-G9
Estimation of Macroscopic Fracture Transport Properties from Micro-Scale Flow Structures and Fracture Geometry
Following the investigation of single-phase flow dynamics in a rough fracture model using computational fluid dynamics, and the early implementation of a modified invasion percolation (MIP) method for two-phase systems (2007 ACS progress report), we continued developing this MIP method to capture capillary drainage, map immiscible fluid location on a discretized, CT-scanned fracture, which lead to the determination of fracture capillary pressure and its correlation structural properties of fractures.
The fracture geometry obtained from CT imaging was used to generate a fracture replica to model capillary-dominated displacement in a fracture with variable apertures and determine capillary pressures by means of a modified invasion percolation (MIP) method. The MIP technique is an algorithm seeking the least resistant pathway for the advancement of the invading fluid using the Young Laplace equation. A quantitative pixel by pixel comparison to experimental CT images validates the applicability of the proposed model. The saturation distribution map generated by MIP yields good agreement with the experimental phase distribution and presents more realistic phase structures than those obtained from the conventional invasion percolation (IP) approach. The developed MIP model was then used to investigate the dependency of fracture capillary behavior on aperture characteristics. A stochastic aperture field generator was employed to create series of artificial fracture models of known geostatistical characteristics. Control parameters of these fracture models were mean aperture, standard deviation, and spatial correlation length. The influence of these three parameters on fracture capillary pressure curves was analyzed and quantified relative to the normalized entry capillary pressure and irreducible water saturation. Results from this investigation allowed us to quantify the effects of fracture morphology on the shape of the fracture capillary pressure curve, and to predict the normalized entry pressure and the irreducible water saturation from global geostatistical parameters describing the fracture aperture field. Findings derived from the most recent stages on this investigation also lead to two publications submitted for peer-review. Each paper emphasizes on a unique aspect of the work. Part I: Model Development and Validation describes the numerical construction of the percolation-based model to simulate immiscible flow in a rough fracture, whose structure had been mapped through high-resolution X-ray CT scanning, the validation of our model through direct comparison against CT images of oil-water distribution, and the determination of fracture capillary pressure curves. Part II: Analysis and Predictions presents extended work with an analysis of the relationship between fracture capillary pressures and geostatistical characteristics of the fracture aperture field, capturing structure, connectivity and tortuosity. Finally, the observed relationships are used to provide practical predictions of fracture capillary pressure indicators. The graduate student assigned to this project, Tawatchai Petchsingto, earned his doctoral degree in May of 2008.
