Untitled

Carmeliet, Jan; Descamps, Filip; Houvenaghel, Geert

doi:10.1023/a:1006500716417

Cited by 76 publications

(13 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As previously mentioned, the prediction of the absorption and drying aspects of a material are important to assess whether a material might be suitable for usage as construction material. These properties are best computed on digital material samples, as not one laboratory techniques covers the entire relevant capillary pressure spectrum [153][154][155] and analyses are often not reproducible between laboratories [12]. Therefore, such unsaturated models could benefit from the developed super-resolution workflow, although additional research is required to also represent the smallest pores because nanoporosity is relevant for construction material studies as well.…”

Section: Super Resolution Models and Their Impact On Fluid Flow Simulmentioning

confidence: 99%

Computed Tomography 3D Super-Resolution with Generative Adversarial Neural Networks: Implications on Unsaturated and Two-Phase Fluid Flow

2020

View full text Add to dashboard Cite

Fluid flow characteristics are important to assess reservoir performance. Unfortunately, laboratory techniques are inadequate to know these characteristics, which is why numerical methods were developed. Such methods often use computed tomography (CT) scans as input but this technique is plagued by a resolution versus sample size trade-off. Therefore, a super-resolution method using generative adversarial neural networks (GANs) was used to artificially improve the resolution. Firstly, the influence of resolution on pore network properties and single-phase, unsaturated, and two-phase flow was analysed to verify that pores and pore throats become larger on average and surface area decreases with worsening resolution. These observations are reflected in increasingly overestimated single-phase permeability, less moisture uptake at lower capillary pressures, and high residual oil fraction after waterflooding. Therefore, the super-resolution GANs were developed which take low (12 µm) resolution input and increase the resolution to 4 µm, which is compared to the expected high-resolution output. These results better predicted pore network properties and fluid flow properties despite the overestimation of porosity. Relevant small pores and pore surfaces are better resolved thus providing better estimates of unsaturated and two-phase flow which can be heavily influenced by flow along pore boundaries and through smaller pores. This study presents the second case in which GANs were applied to a super-resolution problem on geological materials, but it is the first one to apply it directly on raw CT images and to determine the actual impact of a super-resolution method on fluid predictions.Materials 2020, 13, 1397 2 of 33 examples requires detailed knowledge on fluid storage and flow properties which could be estimated through experiments, either carried out in specialized laboratories or deduced from digital rocks. Laboratory measurements, however, require expensive and complex set-ups, take a long time to complete, are often destructive, do not always entirely capture the studied properties and on top of that, experiments often fail to produce reproducible results between laboratories [9,[12][13][14][15][16][17][18][19]. Therefore, numerical solutions have been developed allowing to compute fluid storage and flow properties on digital rocks [7,[20][21][22][23]. Such digital rocks are acquired through direct imaging or numerical simulations of rocks.Computed tomography (CT) is a technique to rapidly capture the three-dimensional structure of materials [22,24,25]. However, like any imaging technique, CT is plagued by a sample size versus resolution trade-off, which is adversely affecting fluid flow simulations. Both a high resolution and sufficiently large volume are required for accurate fluid flow estimations at the pore-scale [9,[26][27][28]. Resolution impacts how well pore network topology and pore surfaces are resolved. At lower resolution, i.e., lower resolving power, pore sizes are often overestimated and specific surf...

show abstract

Section: Super Resolution Models and Their Impact On Fluid Flow Simulmentioning

confidence: 99%

Computed Tomography 3D Super-Resolution with Generative Adversarial Neural Networks: Implications on Unsaturated and Two-Phase Fluid Flow

2020

View full text Add to dashboard Cite

show abstract

“…Other dynamical aspects (behaviour of residual gas pockets related to saturation, prewetting layers, Bosanquet-flow -see Section IV) can also be added [25,32,49,52,63,238,279,281,298]. Simulations can be either quasi-staticthe dynamics is applied at the "weakest link", and no other processes are allowed -or dynamic so that the pressure balance is maintained dynamically at the single pore level [29,30,51,65,74,136,157,238].…”

Section: F Microscopic Simulationsmentioning

confidence: 99%

“…The temporal correlation function (defined in Eq. (49), here the moment q = 2 is considered) could be described by the scaling function with exponents κ = 0.49, θ t = 0.38, and β = 0.63.…”

Section: F Microscopic Simulationsmentioning

confidence: 99%

Imbibition in disordered media

Alava

Dubé

Rost

2004

Advances in Physics

289

333

View full text Add to dashboard Cite

The physics of liquids in porous media gives rise to many interesting phenomena, including imbibition where a viscous fluid displaces a less viscous one. Here we discuss the theoretical and experimental progress made in recent years in this field. The emphasis is on an interfacial description, akin to the focus of a statistical physics approach. Coarse-grained equations of motion have been recently presented in the literature. These contain terms that take into account the pertinent features of imbibition: non-locality and the quenched noise that arises from the random environment, fluctuations of the fluid flow and capillary forces. The theoretical progress has highlighted the presence of intrinsic length-scales that invalidate scale invariance often assumed to be present in kinetic roughening processes such as that of a two-phase boundary in liquid penetration. Another important fact is that the macroscopic fluid flow, the kinetic roughening properties, and the effective noise in the problem are all coupled. Many possible deviations from simple scaling behaviour exist, and we outline the experimental evidence. Finally, prospects for further work, both theoretical and experimental, are discussed.

show abstract

“…[1] constructed a regular two-dimensional pore network model with a presumed pore size distribution to determine sorption isotherms and water vapour permeabilities. [2] improved the method by incorporating measured moisture storage functions as well as permeabilities at low and capillary saturation levels to estimate the unsaturated permeability curve.…”

Section: Introductionmentioning

confidence: 99%

Validation of pore network model for hygric property calculation

et al. 2019

View full text Add to dashboard Cite

Hygric properties can be estimated directly from pore structure information, represented by a network of regularly shaped pores, extracted from a pore structure image to conserve the real topology. On this network, pore-scale models of moisture behaviour determine the hygric properties of moisture storage and transport. The reliability of this approach is validated with a sintered-glass filter. Despite its more limited heterogeneity and pore size range relative to typical porous building materials, it provides a good basis for validating crucial pore-scale moisture processes. Measured storage data compare well to the estimated ad- and desorption moisture retention curves as well as to the saturated and capillary moisture content. Furthermore, the simulated whole-range moisture permeability curve agrees acceptably with measured data. The variation in modelling the pore space as a pore network model is also analysed by considering two distinct pore network extraction methods. The measured and simulated moisture contents agree well for the whole capillary range. Moreover, the resulting transport properties are generally accurate for the whole moisture content range. On the other hand, the estimated vapour permeabilities show notable variations between the two pore network models.

show abstract

Untitled

Cited by 76 publications

References 22 publications

Computed Tomography 3D Super-Resolution with Generative Adversarial Neural Networks: Implications on Unsaturated and Two-Phase Fluid Flow

Computed Tomography 3D Super-Resolution with Generative Adversarial Neural Networks: Implications on Unsaturated and Two-Phase Fluid Flow

Imbibition in disordered media

Validation of pore network model for hygric property calculation

Contact Info

Product

Resources

About