Data-driven auto-tuning strategy for RTO-MPC based on Bayesian optimization
Bibliographic record
Abstract
Real-time optimization (RTO) and model predictive control (MPC) are extensively employed in industrial processes to enhance economic objectives. However, the tuning of the system remains a challenge, in particular, for cases where an explicit model relating the RTO objective and the process dynamics is unknown. To address this issue, an online data-driven auto-tuning strategy leveraging two Bayesian optimization (BO) techniques is proposed. This strategy is useful for situations where the RTO objective and the process dynamics are detectable but their exact functional forms are unclear. The proposed strategy views the RTO objective as a black-box objective and interprets the steady-state conditions as black-box equality constraints within the RTO layer. In this context, the upper-trust-bound based constrained BO (UTB-CBO) method is adopted to optimize the setpoints and enhance solution feasibility. Additionally, the proposed approach can take into account measurable disturbance inputs explicitly and account for their consequential influence on objective optimization by considering disturbance variations as contextual information. Simultaneously, another contextual BO scheme is implemented to automatically tune the MPC controller parameters for improving tracking performance upon accepting the setpoints optimized by the RTO. Simulation results based on a continuous stirred-tank reactor system are given to illustrate the effectiveness of the proposed approach.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".