Simulating secondary waterflooding in heterogeneous rocks with variable wettability using an image‐based, multiscale pore network model

Abstract

The two‐phase flow properties of natural rocks depend strongly on their pore structure and wettability, both of which are often heterogeneous throughout the rock. To better understand and predict these properties, image‐based models are being developed. Resulting simulations are however problematic in several important classes of rocks with broad pore‐size distributions. We present a new multiscale pore network model to simulate secondary waterflooding in these rocks, which may undergo wettability alteration after primary drainage. This novel approach permits to include the effect of microporosity on the imbibition sequence without the need to describe each individual micropore. Instead, we show that fluid transport through unresolved pores can be taken into account in an upscaled fashion, by the inclusion of symbolic links between macropores, resulting in strongly decreased computational demands. Rules to describe the behavior of these links in the quasistatic invasion sequence are derived from percolation theory. The model is validated by comparison to a fully detailed network representation, which takes each separate micropore into account. Strongly and weakly water‐and oil‐wet simulations show good results, as do mixed‐wettability scenarios with different pore‐scale wettability distributions. We also show simulations on a network extracted from a micro‐CT scan of Estaillades limestone, which yields good agreement with water‐wet and mixed‐wet experimental results.

Investigating the relative permeability behavior of microporosity‐rich carbonates and tight sandstones with multiscale pore network models

“…Estaillades has been used frequently as a test case for pore scale modeling and experiments, such as relative permeability studies [Moctezuma-Berthier et al, 2002;Gharbi and Blunt, 2012;Prodanović et al, 2014;Bauer et al, 2012;Ott et al, 2015]. The sample used in this article is the same as the one analyzed in Bultreys et al [2015bBultreys et al [ , 2016b, where our model's relative permeability predictions for drainage and imbibition were compared to experimental data and found to match well.…”

“…Under these circumstances, the results are equally valid to, for example, flow of brine and natural gas, CO 2 , air, or nonaqueous phase liquids. The simulation method is summarized here, full details can be found in Bultreys et al [2015b] (drainage) and Bultreys et al [2016b] (imbibition). The algorithm is based on the method developed by Valvatne and Blunt [2004] but was extended to incorporate fluid transport through the microporosity.…”

Surrogate Models for Studying the Wettability of Nanoscale Natural Rough Surfaces Using Molecular Dynamics

“…In sub-surface fluid reservoirs, such as oil reservoirs, CO 2 storage, or energy storage sites, wettability is known to control the flow efficiency and dynamics [6][7][8]. In computational models, aiming to predict this behavior, the wetting behavior is commonly reflected through contact angles [9,10]. However, conceptually, wettability is a multi-scale thermophysical property, which is the result of a fine balance between fluid-fluid and fluid-solid interactions.…”

The Use of Computed Tomography to Explore the Microstructure of Materials in Civil Engineering: From Rocks to Concrete

“…Porosity, linked to water absorption capacity, is a very important parameter in rocks, as demonstrated by the high number of scientific publications in this field. At present, the most relevant investigations are currently applying CT scan technology to evaluate porosity, regardless of whether it has structural consequences [32,33]. Other investigative works have analyzed mechanical behavior and its connection with the mineralogical microstructure [34].…”