Machine Learning-Additional Decision Constraints for Improved MILP Day-Ahead Unit Commitment Method
Bibliographic record
Abstract
This paper introduces a two-stage (offline and online) artificial neural network (ANN) driven constraint creator model to improve the computational quality of day-ahead unit commitment (DAUC) in power systems. The DAUC is crucial for planning 24-hour operations and complex bid clearing through mixed-integer linear programs (MILP). However, slow convergence is common due to system complexity. Machine learning (ML) based methods have been used to enhance MILP-DAUC. Nonetheless, they can lead to sub-optimality and infeasibility. To overcome these challenges, (1) this paper proposes in the offline stage the ANN-generators subset (AGS) that can predict part of the optimal MILP-DAUC decisions using an ANN model. Online, only ML-generated decisions of AGS are used to form the ANN-driven constraints to enhance the main MILP-DAUC, forming the proposed ANN-MILP-DAUC method. (2)A feasibility handling process is proposed to retain the infeasible ML states to be optimized by the main MILP-DAUC formulation. (3)The proposed model issues an artificial factor that provides the percentage of generators accurately predicted and used as an ML training performance metric. The ANN model was trained using optimal MILP-DAUC solutions. Test results on IEEE 14-bus and 118-bus systems reported solution time reductions of 61.43% and 70.1%, respectively. Larger Polish 2383-bus, 3012-bus, and Ontario systems reported time reductions in the range of 33% compared with the main MILP-DAUC method using MOSEK™, a commercial solver. No degradation in the optimal solution was observed for all test systems, and the proposed method provides a lower-objective solution for the same running time, leading to better solutions.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
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".