Transport in Porous Media最新文献

We present a probabilistic diffusion-based framework for reconstructing scientific microstructure images with missing or corrupted regions. Motivated by challenges in characterizing porous media, our method employs denoising diffusion probabilistic models to learn a conditional distribution over image completions given partial observations. Trained on grayscale images of porous structures, the model generalizes well across samples with varying morphology and entropy. We evaluate its performance on two distinct datasets using a range of masking strategies, including irregular occlusions, large missing regions, and structured patterns such as stripes and cutouts. The proposed model reconstructs high-fidelity completions that are both visually plausible and physically consistent. Quantitative evaluations based on pore size distribution, two-point correlation functions, and pixel-level error metrics show that the generated outputs preserve critical features and statistical descriptors of the original media. Additional analyses of pixel intensity profiles and latent activation patterns reveal that the model can infer fine-scale details while maintaining global structure. These results explain the potential of latent diffusion-based inpainting as a robust tool for digital reconstruction and scientific imaging in complex material systems.

In seasonally frozen regions, ice-water phase transitions during spring snowmelt critically reshape the thermo-hydraulic properties of porous media. However, the underlying pore-scale mechanisms remain poorly quantified, particularly the dynamic variations in thermo-hydraulic transport parameters during melting processes. In this study, a pore-scale numerical model for ice-water phase transitions was developed using the Lattice Boltzmann Method (LBM) with a double-distribution function approach, and its accuracy was rigorously validated. The model enables in-depth investigation of the complex microscopic mechanisms governing coupled heat and fluid flow in porous media under phase change conditions. The results demonstrate that the heterogeneity of porous media structures and thermal boundary conditions jointly govern ice melting dynamics, leading to spatially heterogeneous temperature and phase distributions. Quantitative and qualitative analysis shows that the decrease of porosity significantly speeds up the ice melting rate under the same conditions.

Donald Arthur Nield died peacefully at Golden View Care, Cromwell, New Zealand, surrounded by family, on May 25, 2024, aged 89. He was the dearly loved husband of Rachel, cherished father and father-in-law of Cherry and Robert, Alex and Michael, Peter and Janice, and treasured Grandpa of Elizabeth, John, Charlotte, Frank, Rachel, and Michael. This is a scientific memoir written by D. A. Nield himself. In late March 2017 Don sent a copy of the document, below unedited, to Craig Simmons. This was in response to a discussion that Simmons had with Nield at that time when Simmons was preparing a historical note on the Elder Problem with John W. Elder. We can do no better than to publish Nield’s autobiographical note posthumously as is—in his own words. Nield himself called it “A scientific memoir.”

Graphical abstract

This study investigates the use of X-ray microtomography (XMT) to reveal the structure of complex porous biological tissues and the fluid flow through them during wetting. It also evaluates fluid dynamical simulations based on XMT data to reproduce and analyse these flows, with a final aim of revealing fluid transport and void formation in such tissues. To fulfil the objectives, the wetting flow of a polymer liquid through an initially dry conditioned Norway spruce wood sample is visualised using XMT at the MAX IV synchrotron. The liquid flow front progression captured after 24 s and 48 s reveals uneven filling of longitudinal tracheids and flow between them via the tiny pits which connect tracheids. Most tracheids fill between 24 and 48 s, possibly due to removal of air inclusions. Large density gradients near cell walls suggest that the fluid followed and deposited along wall structures. Computational fluid dynamics simulations (CFD) of saturated flow through the tomography-based geometry indicate velocity profiles that resemble pipe flow in longitudinal tracheids and flow rate differences among them. The latter indicates that the geometry itself may cause the experimentally observed uneven flow. Streamlines show intra-tracheid flow development and clear flow direction change at the pits. Additionally, wetting simulations, using a constant contact angle, capture initial uneven filling between the tracheids on shorter time scales than could be capture by the experiments. These simulations furthermore show air entrapment during filling, consistent with experimental observations. Combining XMT with CFD enables detailed studies of flow in biological porous media. Faster X-ray scanning, incorporating dynamic contact angles and accounting for diffusion in simulations could further refine insights into fluid progression during capillary-driven flow into complex structures of porous biological tissues.

Sequestering carbon dioxide ((hbox {CO}_2)) in deep-ocean sediments is deemed as a promising approach to reducing carbon emissions. Under the low-temperature high-pressure condition of deep-ocean sediments, there may exist hydrate formation zone (HFZ) where solid (hbox {CO}_2) hydrate forms and negative buoyancy zone (NBZ) where (hbox {CO}_2) is denser than water. Both of the HFZ and the NBZ suppress the upward movement of the (hbox {CO}_2) plume; therefore, permanent storage was proposed in the deep-ocean sediment even if there is no low-permeability caprock on the top of the reservoir. However, in virtue of numerical simulations on (hbox {CO}_2) injection over a wide range of deep-ocean sediment conditions, we find that neither the HFZ, the NBZ nor the combination of the HFZ and NBZ makes sufficient condition for permanent (hbox {CO}_2) storage in the deep-ocean sediment, and we cannot evaluate the (hbox {CO}_2) storage security simply based on the existence of the HFZ and the NBZ. This is because (1) only a very small amount of hydrate forms in the HFZ and the formed hydrate may dissociate with continuous (hbox {CO}_2) injection and (2) the negative gravitation trapping by the NBZ may fail if the permeability of the sediment is not high enough to make the negative buoyancy force effective. We also find that the NBZ may shrink because the temperature increase due to exothermic hydrate formation may significantly reduce (hbox {CO}_2) density and we propose a new method to calculate the size of the NBZ. Finally, unconditional permanent (hbox {CO}_2) storage may only exist in high-permeability sediments with NBZ.

Structure and dynamics of turbulent open channel flow over permeable and impermeable sediment beds are investigated using pore-resolved, direct numerical simulations. Time-space double-averaged statistics are computed in four configurations: (i) permeable bed with randomly packed sediment grains, (ii) an impermeable wall with full layer of roughness elements matching the top layer of the sediment bed, (iii) an impermeable wall with half layer of roughness elements, and (iv) a smooth wall. It is observed that the mean velocity, Reynolds stresses, and form-induced pressure–velocity correlations representing ejection and sweep fluxes are similar in magnitude for the permeable-bed and impermeable full-layer cases. The wall-blocking effect present in the impermeable half layer results in higher streamwise and lower wall-normal stresses compared to the permeable bed. Bed roughness increases Reynolds shear stress, whereas permeability has minimal influence. However, bed permeability significantly influences form-induced shear stress. Pressure fluctuations and volume-averaged bed-normal distribution of the drag force peak in the top layer of the bed. These findings suggest that reach-scale transport in the hyporheic zone will be better captured by providing boundary conditions based on stream flow simulations that incorporate the roughness effect of the top layer of the bed.

This study investigates the impact of fin parameters on natural convection heat transfer in the closed cavity with a heat source. The analysis focuses on the comparison of the thermal performance between solid and porous fins. Firstly, the influence of individual parameter changes is analyzed. Based on these findings, the response surface optimization method is applied to explore the heat transfer characteristics when multiple fin parameters vary simultaneously. The results of single parameter variation show that the installation angle of porous fins has the most significant influence on the average Nusselt number of the heat source surface. For solid fins, the fin length has the greatest impact. The interaction between the installation angle and the length of the porous fin has the most significant effect on the average Nusselt number of the heat source surface, reaching a maximum value of 11.65. Compared to the cavity without fins, the optimal configuration enhances the average Nusselt number by 12.02%. The corresponding optimal parameters for the porous fin are θ = 118.3°, l = 0.288H and a = 0.664H. Similarly, for the solid fin, the interaction between the installation angle and fin length has the most significant effect on the average Nusselt number of the heat source surface. Correspondingly, the maximum average Nusselt number on the surface of the heat source reaches 11.57, representing an 11.32% increase compared to the cavity without fins. The optimal parameters for the solid fin are θ = 108.9°, l = 0.021H, a = 0.747H.

In this study, we analyze combustion waves in a simplified one-dimensional model for porous media, focusing on the case of backward propagating combustion where the combustion front moves opposite to the direction of the injected airflow, resulting in negative wave velocity. The mathematical model consists of three coupled partial differential equations governing temperature, oxygen, and fuel concentrations. Assuming that oxygen is transported faster than heat, we reduce the system to a form suitable for phase plane analysis and establish the existence of counterflow traveling wave solutions. Our work extends previous results on coflow combustion waves by providing a comprehensive classification of counterflow wave types and their properties. The existence and structure of these waves are rigorously demonstrated through dynamical systems techniques, offering new insights into the behavior of combustion fronts in porous media.

In this work, the modelling of entropy generation on heat transport of natural convection of an electrically conducting Walters’ B fluid is examined. The flow through a porous medium radiates nonlinearly in the presence of viscous dissipation and Joule heating. Subject to the suitable dimensionless variables, the coupled nonlinear dimensional equations are transformed into ordinary differential equations via a similarity variable and executed by Galerkin Weighted Residual Method (GWRM). The results obtained demonstrated good agreement with another method when validated by Spectral Collocation Method (SCM) through tables, as well as numerical integration of Mathematica’s NDSolve for the graphs. The dynamics of the embedded parameters are presented through graphs. Keeping in mind the engineering applications of the study, the Skin-friction and Nusselt number results are conveyed through tables. The result justified, among other important findings, that temperature distribution cools over the higher dominance of buoyancy force over the viscous force, which is a useful tool in application for cooling of the system. The interaction of the Brinkman number intensifies viscous heating due to the heat transfer by virtue of the molecular conduction around the system. The outcome of this process improves entropy production in applications.