Enhancing Safety in Geological Carbon Sequestration: Supervised Machine Learning for Early Detection and Mitigation of CO2 Leakage in Injection Wells
Bibliographic record
Abstract
Abstract The efficient and safe operation of CO2 injection wells during geological sequestration is crucial for successful carbon capture and storage (CCS) projects. This study explores the application of machine learning in creating a data-driven model for simultaneous prediction of the location and size of potential leak incidents along an active CO2 injection well based on wellhead and bottom-hole pressure and temperature data. Five different well-established machine learning algorithms were selected for predictive model development, including Support Vector Regression (SVR), K-Nearest Neighbor Regression (KNNR), Decision Tree Regression (DTR), Random Forest Regression (RFR), and Artificial Neural Network (ANN). A series of numerical simulations were performed to create a dataset based on a CO2 injection well model in a southern North Sea saline aquifer reservoir, accounting for various leak scenarios with different locations and sizes. The dataset includes three input features of wellhead pressure, bottom-hole pressure, and bottom-hole temperature, paired with two output variables of leak location and leak size. The research findings demonstrate that all models perform well in effectively pinpointing leak locations, but they face difficulties when it comes to detecting small leaks, particularly those with a CO2 leakage rate below 0.01 kg/s. The results obtained indicated that, with regard to model performance, the SVR and KNNR models tended to outperform the others during the testing phase. More precisely, the SVR model demonstrated exceptional performance in the context of leak localization, particularly when dealing with smaller datasets. Conversely, KNNR consistently showcased superior performance in the detection of leak size, regardless of the dataset size. The outcomes of this research can provide valuable insights into the behavior of leaky CO2 injection wells during geological sequestration and highlight the efficacy of supervised machine learning in detecting and predicting leakage in CO2 injection wells.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), 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".