Differential Evolution for Assisted History Matching Process: SAGD Case Study
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
Abstract SAGD (Steam Assisted Gravity Drainage) is an efficient and proven technology to recover vast reserves of Alberta's oil sands. Because of its thermal and compositional effects, numerical simulation of the SAGD process requires extensive computational run time, especially in a history matching framework. Therefore, it is beneficial to use an optimization technique that yields faster convergence and better match-quality solutions. This paper presents a new population-based optimization technique, called differential evolution, in the assisted history matching process. Differential evolution belongs to the class of evolutionary algorithms in the continuous parameter space that has been used successfully in a large range of engineering optimization problems outside the oil industry. Differential evolution converges faster than many other global optimization methods. It requires fewer control variables, is robust and easy to use, and lends itself very well to parallel computing. We applied the differential evolution technique to a SAGD case study to history match saturation and temperature profiles as well as cumulative oil and water production and cumulative SOR. The results show that it is an excellent optimization technique for obtaining multiple good history matched models, which allow the assessment of uncertainty for the forecast stage. The match-quality of the history matched models obtained with differential evolution has been compared to the results of the particle swarm optimization method that is widely used in history matching. The comparison shows that differential evolution offers much better match-quality solutions with much lower number of simulation runs.
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 | 0.000 |
| 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 it