Application of machine learning in modelling gas dispersion coefficients for hydrogen storage in porous media
Bibliographic record
Abstract
Abstract Underground hydrogen storage (UHS) in geological formations is recognized as a promising solution for managing renewable energy intermittency and supporting large‐scale energy systems. However, the mixing of hydrogen with cushion gases such as methane (CH 4 ), nitrogen (N 2 ), and carbon dioxide (CO 2 ) remains a critical challenge, directly affecting storage efficiency and gas purity. The gas dispersion coefficient (KL) plays a fundamental role in governing this mixing behaviour, yet its accurate prediction under realistic reservoir conditions has been limited in previous studies. This research presents a novel approach by integrating core flooding experiments with advanced machine learning (ML) techniques to estimate KL values with high precision. Unlike earlier studies that primarily relied on analytical models or limited experimental data, this work combines systematic ML modelling with extensive laboratory data to capture the complex, nonlinear nature of gas dispersion in porous media. The results indicate that support vector regression (SVR) provides superior predictive performance for all tested gases. Specifically, for CH 4 , N 2 , and CO 2 , the SVR model achieved coefficient of determination ( R 2 ) values of 0.9968, 0.9977, and 0.9973, respectively, along with low mean absolute deviation (MAD) values of 0.014, 0.008, and 0.013, and root mean square error (RMSE) values of 0.017, 0.011, and 0.016. These findings provide valuable insights for optimizing cushion gas selection and improving the accuracy of UHS system design, ultimately enhancing storage reliability and contributing to more efficient and sustainable large‐scale hydrogen storage.
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.
How this classification was reachedexpand
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 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".