Safe optimization with grey-box information: Application to composites autoclave processing improvement on the fly
Bibliographic record
Abstract
In the manufacture of aerospace-grade composites in the autoclave, the curing process plays a crucial role as it directly governs the quality of the final parts. Maintaining the part’s thermal history, namely, thermal lag and exotherm, under predetermined thresholds as well as achieving a uniform degree of cure throughout the material thickness can result in the desired product quality. Currently, for many such manufacturing applications, the optimization of the curing process (often conducted via trial-and-error) is highly expensive and time-consuming and occasionally leads to failed products. In order to address this problem, in this paper, a Safe Optimization approach is proposed. The suggested framework allows for the on-the-fly optimization of curing process configurations while avoiding interruptions typically encountered during trials. In other words, the proposed algorithm is capable of consistently yielding “pass” products as it navigates toward the optimal configuration. In particular, we introduce a hybrid optimization framework that combines a genetic algorithm, namely NSGA-II, using inexpressive stimulation (white-box) data for finding a safe initial starting point and then, the (black-box) safe logarithmic barrier method for enhancing the product quality presumably using experimental data on-the-fly. Herein, however, as proof of concept, we employ synthetic data throughout the framework in a case study.
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 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".