A computationally efficient modeling of flow in complex porous media by coupling multiscale digital rock physics and deep learning: Improving the tradeoff between resolution and field-of-view

Advances in Water Resources, Journal Year: 2024, Volume and Issue: 188, P. 104695 - 104695

Published: April 9, 2024

Digital rock physics is at the forefront of characterizing porous media, leveraging advanced tomographic imaging and numerical simulations to extract key properties like permeability. However, fully capturing heterogeneity natural rocks necessitates increasingly larger sample volumes, presenting a significant challenge. Direct these scales become either prohibitively expensive or computationally unfeasible due limitations in resolution field view (FOV). This issue particularly pronounced carbonate rocks, known for their complex, multiscale pore structures, which exacerbate resolution-FOV tradeoff. To address this, we introduce machine learning strategy that merges data from various resolutions with 3D convolutional neural network (CNN) model. approach innovative its ability identify cross-scale correlations, thereby enabling estimation transport volumes—properties are difficult simulate directly—using trainable proxies. The integration deep allows accurate permeability predictions beyond those feasible traditional direct simulation methods. By employing transfer across different during training phase, our model achieves robust performance, an R² exceeding 0.96 when evaluated on diverse lower-resolution domains FOVs. Notably, this method significantly enhances computational efficiency, reducing time by orders magnitude. Originally developed intricate structures shows promise application wide range offering viable solution longstanding tradeoff between FOV digital physics.

Language: Английский

---

Mechanical properties and failure mechanism of 3D printing ultra-high performance concrete

Construction and Building Materials, Journal Year: 2024, Volume and Issue: 447, P. 138108 - 138108

Published: Aug. 30, 2024

Language: Английский

---

Digital rock super-resolution reconstruction with efficient 3D spatial-adaptive feature modulation network

Geoenergy Science and Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 213748 - 213748

Published: Feb. 1, 2025

Language: Английский

---

Assessment of deep-learning-based resolution recovery algorithm relative to imaging system resolution and feature size

Methods in microscopy, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 10, 2025

Abstract High-resolution X-ray microscopy is crucial for non-destructive materials characterization, but achieving both high resolution and maintaining a wide field of view often necessitates time-consuming approaches. Deep learning super-resolution methods based on convolutional neural networks are bridging this gap to obtain high-resolution usable data analysis from low-resolution images. This study evaluates novel deep learning-based algorithm designed overcome traditional limitations by spatially varying point spread function set registered low- image pairs. With systematic methodology, we evaluated the algorithm’s superior performance in recovering features across range resolutions with increasing quality degradation. It was also benchmarked against classical iterative Richardson-Lucy deconvolution algorithm, well-known deep-learning-based network SRCNN same Qualitative quantitative evaluations using simulated foam phantoms showed that our shows excellent feature recovery, within 5 % ground truth, even large ratio 7:1 between high- Multiscale investigations real porous material semiconductor device presented highlight its recovery versatility real-world scenarios.

Language: Английский

---

Quantifying reservoir microfacies characterization using thin-section, scanned, computed tomography, and electron microscope image data

Elsevier eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 329 - 360

Published: Jan. 1, 2025

Language: Английский

---

Super-resolution enhancement of X-ray microscopic images of solder joints

NDT & E International, Journal Year: 2025, Volume and Issue: unknown, P. 103382 - 103382

Published: March 1, 2025

Language: Английский

---

Geosystems risk and uncertainty: The application of ChatGPT with targeted prompting

Geoenergy Science and Engineering, Journal Year: 2024, Volume and Issue: 238, P. 212889 - 212889

Published: May 9, 2024

ChatGPT, a prominent large language model (LLM), is being increasingly used across wide range of scientific fields. Geosystem engineers and researchers are also posed to leverage ChatGPT find solutions challenges encountered in various topics. This study evaluates the accuracy reproducibility responding different qualitative quantitative questions, with particular focus on risk uncertainty (R&U) both Greenfield Brownfield domains as an important area interest. The results show importance prompting considerably improve ChatGPT's response reproducibility. For example, increases responses questions domain by 10.4% 41.8%, respectively. Additionally, enhances responses, 32.1% increase for 33.3% rise domain. findings highlight that greater comprehensiveness prompts, higher questions. acknowledges potential limitations associated sources information contextual influences reliability response.

Language: Английский

---

Application of Hierarchical Homogenization Technique in Thermal Conductivity Computation for Micro Computed Tomography (Micro-CT) Images of Porous Media

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: March 18, 2025

Thermal properties play a critical role in environments and processes involving heat transfer. Heat transport porous media has been subject of extensive study due to its significant impact on applications ranging from situ hydrocarbon production geothermal energy projects. Micro-CT imaging emerged as an innovative technique for characterizing media, with adoption growing recent years advancements computational methods. Homogenization approaches provide powerful means analyze phenomena images, offering reliable accuracy while reducing errors. In this study, the application Hierarchical (HH) thermal conductivity was explored. Various sources error, including choice homogenization scale numerical conditions such padding thickness, were systematically investigated. The results indicated less than 5% error first order single stage HH approach all studied material schemas. Hyperbolic trend observed homogenization. Subsequently, Telescopic (THH) found effective new more complex systems negligible (less %1.5) compared HH. Furthermore, investigated set 19 synthetic real samples assess effect porosity variation sub final values mathematical model obtained each Results showed that similar cases, sample higher dispersion will result through procedure.

Language: Английский

---

Pore-scale fluid distribution and remaining oil during tertiary low-salinity waterflooding in a carbonate

Petroleum Science, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 1, 2024

Language: Английский

---

Numerical Simulation Analysis of Control Factors on Acoustic Velocity in Carbonate Reservoirs

Minerals, Journal Year: 2024, Volume and Issue: 14(4), P. 421 - 421

Published: April 19, 2024

The conventional Archie formula struggles with the interpretation of water saturation from resistivity well log data due to increasing complexity exploration targets. This challenge has prompted researchers explore alternative physical parameters, such as acoustic characteristics, for breakthroughs. Clarifying influencing factors porous media characteristics is one most important approaches help understanding mechanism carbonate reservoirs. article uses digital rock technology characterize pore structure, quantitatively identify fractures and structures in rocks, establish models. Through testing, pressure wave (P-wave) shear (S-wave) velocities samples at different saturations are obtained, dynamic elastic modulus calculated. A finite element calculation model established using computational provide a basis fluid methods. Based on real models, combinations virtual constructed, affecting parameters analyzed. study finds that porosity increases, velocity difference between cores fractured also increases. These findings technical support theoretical interpreting logging evaluating reservoirs fracture types.

Language: Английский

---