Development of artificial neural networks (ANNs) for chemistry representation in the conditional source-term estimation (CSE) combustion model
Bibliographic record
Abstract
The objective of this study is to develop and implement artificial neural networks (ANNs) with real-time integration in Conditional Source-term Estimation (CSE) used for turbulent combustion modelling. Aspects related to prediction accuracy, storage and computational cost are examined. For the first time, ANNs will be coupled with CSE to determine conditional averages of temperature, species mass fractions and source terms. Two sets of ANNs are developed for two different non-premixed turbulent flames: a pure methane jet flame and a diluted methane jet flame. The ANNs are trained and tested with augmented tabulated chemistry data. Reasonable accuracy is obtained during the testing process for both sets of ANNs across all mixture fractions, and a storage reduction of over 66% is obtained for both fuels. The CSE routine is then modified to replace the tabulated chemistry search and interpolation process with the ANN calculations (ANN-CSE), which is then used to simulate both flames in a Reynolds-Averaged Navier-Stokes (RANS) framework. ANN-CSE is able to produce reasonable results for the Favre averages of species mass fractions and source terms for both flames. The largest conditional deviations between ANN-CSE and conventional CSE with tabulated chemistry results occur in the fuel-rich region in both flames, the differences are reduced in the Favre averages. Further, ANN-CSE requires about 24% less memory than CSE with tabulated chemistry. The computational time is reduced by over 44% for both flames. ANNs are a promising method for representing complex fuel chemistry in the CSE combustion model and can be extended to other fuels and combustion regimes.
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 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".