Integration of machine learning and data analysis for the <scp>SAGD</scp> production performance with infill wells
Bibliographic record
Abstract
Abstract There have been numerous studies on predicting the production performance of the steam assisted gravity drainage (SAGD) process by data‐driven models with different machine learning algorithms since their introduction into industry. Similar efforts on SAGD infill wells, nevertheless, remain rare for this advanced alteration in improving the classical SAGD performance. On the other hand, predictive tools to optimize an infill well start time is useful in maximizing bitumen production and minimizing its costs. In this paper, a series of SAGD infill well models are constructed with selected ranges of operational conditions. Three SAGD infill well production performance indicators, namely, an increased ratio (), a total steam–oil ratio (SOR total ), and a stolen ratio () for each SAGD infill well, are calculated based on simulated infill well cases and control models. Five different machine learning algorithms (an artificial neural network [ANN] algorithm, three gradient boosting decision tree [GBDT] algorithms, and a support vector machine [SVM] algorithm) are trained, tested, and evaluated for their effectiveness in predicting those three indicators as output parameters, given seven SAGD relevant parameters as input parameters. Comparisons of different data sets show that the ANN is the best in predicting all three performance indicators under different infill well start times among all the above machine learning algorithms, while the GBDT algorithms have a better ability to learn a variation trend in the SAGD infill well performance.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| 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) | 0.001 | 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".