Enhancing Effective Thermal Conductivity Predictions in Digital Porous Media Using Transfer Learning
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Porous media beneath the Earth’s surface, including aquifers, oil and gas reservoirs, and geothermal systems, play a crucial role in various natural resource management and environmental engineering applications. The study of their physical properties, particularly thermo-physical properties like effective thermal conductivity (ETC), is essential for enhancing the efficiency of subsurface engineering technologies including nuclear waste disposal, geothermal energy utilization, and underground thermal energy storage. Traditionally, determining ETC has relied on either simplified empirical models, which often lack accuracy, or sophisticated laboratory experiments, which are time-consuming and resource intensive. The advent of three-dimensional (3D) imaging technologies has enabled digital characterization of subsurface media, but direct numerical simulations of ETC remain computationally prohibitive. In response to these challenges, we introduce a novel machine learning framework that leverages transfer learning to enhance the prediction of ETC in digital rock samples. Our approach utilizes state-of-the-art convolutional neural networks (CNNs), pre-trained on extensive datasets, and applies them to various porous media samples, including Berea sandstone, Bentheimer sandstone, and Ketton limestone. By employing transfer learning, we demonstrate that our models can achieve high prediction accuracy with significantly reduced training time, computational power, and data requirements. This study highlights the potential of transfer learning to advance the efficiency and accuracy of digital rock analysis, offering a promising tool for the rapid and reliable characterization of subsurface properties.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it