A hybrid CFD-neural network framework for the early prediction of longitudinal thermo-acoustic instabilities in hydrogen-fueled gas turbine combustors
Bibliographic record
Abstract
This study introduces a hybrid Computational Fluid Dynamics–Neural Network (CFD–NN) framework for real-time prediction of longitudinal thermo-acoustic instabilities in hydrogen-fueled gas turbine combustors—critical for future clean energy systems. A high-fidelity two-dimensional CFD model simulated hydrogen combustion and provided time-resolved pressure and heat release data. The Local Rayleigh Index (LRI) identified probe X 2 as a strong instability zone (LRI ≈ 10⁷ Pa·W), while X 1 , X 4 , and X 6 showed stable behavior. A feedforward neural network trained on early-stage data from probe X3 achieved high prediction accuracy (R² = 0.9998, Root Mean Square Error (RMSE = 924)) and delivered predictions ∼676× faster than CFD (∼334 predictions/s). By combining physics-based modeling with machine learning, this hybrid method enables real-time, physics-informed diagnostics, supporting smart combustor design and closed-loop control in next-gen hydrogen turbines. • A hybrid CFD–FNN framework is proposed for early prediction of thermoacoustic instability in hydrogen-fueled gas turbines. • High-fidelity 2D CFD extracted pressure and heat-release data at six probes, using the Rayleigh Index to identify instability. • Probe X3 enabled earliest instability detection, while X5 captured transitional dynamics supporting mid-stage prediction. • An FNN trained on probe X3 achieved R² = 0.9998, low RMSE, and delivered a 676× speed-up over CFD for real-time prediction. • The framework enables physics-informed real-time diagnostics, supporting closed-loop combustion control in hydrogen turbines.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".