FFT-based deep learning for efficient combustion instability prediction: A comparative study of time and frequency-domain approaches
Bibliographic record
Abstract
This study presents a deep learning-based approach for real-time prediction of combustion instability. Two LSTM models were developed: one trained on time-series data of pressure and heat release rate (OH* intensity), and another on frequency-domain features derived via fast Fourier transform (FFT). The dataset includes measurements taken under varying power levels (15–30 kW), hydrogen content (0%–80%), air flow rates (400–600 slpm), and downstream acoustic conditions with blockage ratios of 0, 0.73, and 0.85. Both models demonstrated high accuracy within a 100 ms window, 93.67% for the time-series model and 95.11% for the FFT-based model. However, the FFT-based model achieved 4.8 × faster inference, making it more suitable for real-time deployment. Even at smaller windows, it maintained comparable accuracy (94.28% vs. 94.76%). Additionally, both models were tested on transitional regimes (stable to unstable and vice versa) labeled using the Rayleigh Index. The models showed strong alignment with these transitions, particularly for unstable-to-stable cases, confirming their reliability in dynamic operating conditions. • Developed LSTM-based models to predict combustion instability in hydrogen- enriched flames. • Compared time-series and FFT-based approaches using pressure and OH* signals. • FFT-based model achieved comparable accuracy (95.11%) with 4.8 × faster inference speed. • Validated on transitional regimes labeled via Rayleigh Index for reliability. • Offers an efficient tool for real-time hydrogen combustion monitoring.
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 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| 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".