Distributed Real-Time Multi-Pad Steam Allocation Optimization
Bibliographic record
Abstract
Abstract In Steam-Assisted Gravity Drainage (SAGD) recovery, optimal real-time steam allocation from a shared steam generator to the physically coupled multi-pads can significantly improve long-term performance goals. However, multi-pad real-time optimization (RTO) with first-principle models can be computationally intensive. Furthermore, general-purpose optimization algorithms in RTO do not consider the future state beyond the prediction horizons to be optimized and treat the optimization problem as a long-term optimization process. Since steam is the primary cost factor in SAGD, Key Performance Indicators (KPI) such as Net Present Value (NPV), when used in RTO, result in low steam injection impeding steam chamber growth during the build-up and normal SAGD operational phase. Therefore, balancing steam chamber development and economics becomes essential for SAGD well-pads using RTO to meet long-term goals. In this contribution, we implement the Alternating Direction Method of Multipliers (ADMM) and a dynamic data-driven model to reduce the computational cost of RTO. ADMM coordinates in real-time field-wide use of shared steam generation. The shared steam generation is a market commodity traded between the pads, with global coordination in real-time perturbation of their market prices. Four SAGD KPIs are implemented for a multi-pad RTO of the SAGD normal operations phase to see which KPI eventually grows the steam chamber without negatively affecting the long-term economic performance. A SAGD field with four pads with 33 well-pairs shows that for all four pads, an economic-based KPI limits the achievement of long-term goals because it cannot account for the future state beyond the horizon under consideration due to hindered steam chamber growth. For the steam chamber expansion and bitumen recovery KPI, high recovery and economic performance are achieved, but with a high resource requirement, leading to a high carbon footprint. On the other hand, an alternating economic and bitumen recovery KPI achieves high economic performance while minimizing resource requirements that decrease carbon footprint.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 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".