Comparison Study of Two Different Methods on the Localised Enkf on SAGD Processes
Bibliographic record
Abstract
Abstract The localization on the automatic history matching of SAGD processes has not been studied. Also the distance-based localization is not applicable to SAGD processes. That is because in SAGD processes oil is produced mainly from the transition zone instead of regions centered producer. Distance cannot be used as the localization scale for SAGD processes. As the transition zone could be defined by temperature distribution, a new localization function with temperature as localization scale was developed. The temperature-based localization function was obtained through modifying distance-based localization function. The localization regions were determined through covariance analysis by using a large ensemble. Based on the covariance analysis, the temperature range of the transition zone is determined. The localization function is 1.0 for the regions within this temperature range. Beyond or below this range, the localization function reduces from 1.0 and at the critical or steam temperature the localization function reduces to zero. The localization is applied to covariance of data with permeability, saturation and temperature, as well as the covariance of data with data. The general localization function was developed by sensitivity analysis with synthetic cases. It could be applied to any SAGD cases with different steam temperatures and reservoir parameters. In the numerical simulation process, the history matching results showed that without localization, the variability in the ensemble collapsed very quickly and lost the ability to assimilate later data with a small ensemble (50 ensemble members) while the prediction was far from the reference with data mismatch remaining at a high level. The temperature-based localization is able to avoid the collapse of the ensemble variability with a smaller ensemble (10 ensemble members) which saves computation time and gives better history-match and prediction results. The computation time was analyzed in the work. With the temperature-based localization the computation time is greatly decreased by 75%. And another approach (localised EnKF with oil saturation) is also applied in this study to compare the calculation rate and accuracy. For this two methods, the localization regions should be the regions that have true correlations with production data. Dynamic data-temperature and oil saturation is used as localization scale in the localization function. And the temperature range and oil-saturation range is defined at which oil is flowing. And critical temperature is defined at which oil begins to get mobile. The localization function is 1 for regions in the temperature and oil saturation range, and decreases to 0 when the temperature reaches the steam temperature or critical temperature, or oil saturation reached initial oil saturation or residual oil saturation. To get the best history matching and prediction results, the localization function is applied to the Kalman gain and covariance matrices separately.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| 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".