Physically Interpretable AI-Driven Prediction of Instability Regimes in Hydrogen-Enriched Premixed Combustion
Abstract
Hydrogen-enriched combustion is central to decarbonizing high-efficiency energy systems, yet its practical adoption is limited by the onset of thermoacoustic and hydrodynamic instabilities in premixed flames. In this study; a novel, integrated framework that combines high-fidelity computational fluid dynamics (CFD) simulations with interpretable deep learning for the prediction and physical diagnosis of combustion instability was proposed. A parametric suite of 1,500 axisymmetric CFD simulations was carried out, systematically varying hydrogen blending ratios (0–100% by volume), equivalence ratios (ϕ = 0.6–1.4), and turbulence intensities (5–25%). Key instability markers including root-mean-square (RMS) pressure, flame front wrinkling, and radical pool dynamics were extracted from both stable and unstable flame regimes. The data collected was used to train a hybrid convolutional neural network–long short-term memory (CNN–LSTM) model, which achieved a test accuracy of 94.3%, F1-score of 94.4%, and area under the receiver operating characteristic curve (AUC-ROC) of 0.978 in binary regime classification. SHAP-based interpretability analysis demonstrated that the model’s predictions were grounded in physically relevant features, with RMS pressure, OH fluctuations, and dominant acoustic frequencies serving as the principal contributors. AI-predicted instability regime maps showed an 88.6% overlap with CFD-derived instability thresholds, highlighting the physical consistency of the approach. Distinct field visualizations showed that unstable regimes (ϕ = 1.1, H₂ = 80%) exhibit pronounced front wrinkling, broader high-temperature zones, and spatially distributed radical production compared to stable flames. This approach opens a promising path for data-driven, physically interpretable instability diagnostics, which could directly impact for burner design, operational safety, and real-time combustion monitoring in hydrogen-based systems. In future work, it is aimed to extend this approach to multi-fuel configurations and experimental integration for real-world deployment.
Keywords
Hydrogen-enriched combustion, Premixed flame instability, CFD simulation, AI-based combustion regime classification
Supporting Institution
This work was supported by Cukurova University, Department of Scientific Projects (Project no: FBA-2024-16686).
Ethical Statement
This study does not involve human or animal participants. All procedures followed scientific and ethical principles, and all referenced studies are appropriately cited.
Thanks
The author declare that there are no acknowledgements.
References
- Abdel-Fattah, A., Wahba, E., & Mahrous, A.-F. (2025). Experimental and numerical study of turbulent fluid flow of jet impingement on a solid block in a confined duct with baffles. International Communications in Heat and Mass Transfer, 160, Article 108294. https://doi.org/10.1016/j.icheatmasstransfer.2024.108294
- Agostinelli, P. W., Laera, D., Chterev, I., Boxx, I., Gicquel, L., & Poinsot, T. (2022). On the impact of H2-enrichment on flame structure and combustion dynamics of a lean partially-premixed turbulent swirling flame. Combustion and Flame, 241, Article 112120. https://doi.org/10.1016/j.combustflame.2022.112120
- Beita, J., Talibi, M., Sadasivuni, S., & Balachandran, R. (2021). Thermoacoustic instability considerations for high hydrogen combustion in lean premixed gas turbine combustors: A review. Hydrogen, 2(1), 33–57. https://doi.org/10.3390/hydrogen2010003
- Berger, L., Attili, A., & Pitsch, H. (2022). Intrinsic instabilities in premixed hydrogen flames: Parametric variation of pressure, equivalence ratio, and temperature. Part 1—Dispersion relations in the linear regime. Combustion and Flame, 240, Article 111935. https://doi.org/10.1016/j.combustflame.2021.111935
- Cellier, A., Lapeyre, C. J., Oztarlik, G., Poinsot, T., Schuller, T., & Selle, L. (2021). Detection of precursors of combustion instability using convolutional recurrent neural networks. Combustion and Flame, 233, Article 111558. https://doi.org/10.1016/j.combustflame.2021.111558
- Chen, X., Liu, Q., Jing, Q., Mou, Z., Shen, Y., Huang, J., & Ma, H. (2022). The characteristics of flame propagation in hydrogen/oxygen mixtures. International Journal of Hydrogen Energy, 47(17), 10069–10082. https://doi.org/10.1016/j.ijhydene.2022.01.097
- Cheng, J., Liu, B., & Zhu, T. (2024). Effects of fuel/air mixing distances on combustion instabilities in non-premixed combustion. Physics of Fluids, 36(7), Article 074113.
- Driscoll, J. F. (2008). Turbulent premixed combustion: Flamelet structure and its effect on turbulent burning velocities. Progress in Energy and Combustion Science, 34(1), 91–134. https://doi.org/10.1016/j.pecs.2007.04.002
- Echekki, T., & Mastorakos, E. (Eds.). (2011). Turbulent combustion modeling: Advances, new trends and perspectives. Springer. https://doi.org/10.1007/978-94-007-0412-1
- Gal, Y., & Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In Proceedings of the 33rd International Conference on Machine Learning (pp. 1050–1059). PMLR.