Hidrojenle Zenginleştirilmiş Ön Karışımlı Yanmada Kararsızlık Rejimlerinin Fiziksel Olarak Yorumlanabilir Yapay Zekâ Destekli Tahmini

Abstract

Hidrojenle zenginleştirilmiş yanma, yüksek verimli enerji sistemlerinin karbonsuzlaştırılmasının merkezinde yer alır; ancak ön karışımlı alevlerde ortaya çıkan termoakustik ve hidrodinamik kararsızlıklar, pratik uygulamadaki yaygınlaşmasını sınırlamaktadır. Bu çalışmada, yanma kararsızlığının öngörüsü ve fiziksel tanılanması için yüksek doğruluklu hesaplamalı akışkanlar dinamiği (CFD) simülasyonlarını yorumlanabilir derin öğrenme ile birleştiren yeni ve entegre bir çerçeve önerilmiştir. Eksensimetrik 1500 CFD simülasyonundan oluşan parametrik bir dizi yürütülmüş; hidrojen karışım oranları (hacimce %0–100), eşdeğerlik oranları (ϕ = 0.6–1.4) ve türbülans şiddetleri (%5–%25) sistematik olarak değiştirilmiştir. Kök-ortalama-kare (RMS) basınç, alev önü buruşması ve radikal havuzu dinamiği dâhil temel kararsızlık göstergeleri, hem kararlı hem de kararsız alev rejimlerinden çıkarılmıştır. Toplanan veriler, hibrit bir evrişimsel sinir ağı–uzun-kısa süreli bellek (CNN–LSTM) modelini eğitmek için kullanılmış; model, ikili rejim sınıflandırmasında %94.3 test doğruluğu, %94.4 F1-skoru ve 0.978 AUC-ROC değeri elde etmiştir. SHAP tabanlı yorumlanabilirlik analizi, model tahminlerinin fiziksel olarak anlamlı özniteliklere dayandığını; RMS basınç, OH dalgalanmaları ve baskın akustik frekansların başlıca katkı sağlayanlar olduğunu göstermiştir. Yapay zekâ ile öngörülen kararsızlık rejim haritaları, CFD’den türetilen kararsızlık eşiklerinin %88.6’sı ile çakışarak yaklaşımın fiziksel tutarlılığını ortaya koymuştur. Alan görselleştirmeleri, kararsız rejimlerin (ϕ = 1.1, H₂ = %80) kararlı alevlere kıyasla belirgin alev önü buruşması, daha geniş yüksek sıcaklık bölgeleri ve uzamsal olarak dağıtılmış radikal üretimi sergilediğini göstermiştir. Bu yaklaşım, veri odaklı ve fiziksel olarak yorumlanabilir kararsızlık tanılaması için umut verici bir yol açmakta; brülör tasarımı, işletme güvenliği ve hidrojen temelli sistemlerde gerçek zamanlı yanma izlemesine doğrudan etki potansiyeli taşımaktadır. Gelecek çalışmalarda, yöntemin çoklu yakıt konfigürasyonlarına ve gerçek dünya uygulamaları için deneysel entegrasyona genişletilmesi hedeflenmektedir.

Keywords

• Hidrojence zengin yanma
• Ön yanma alev kararsızlığı
• CFD simülasyonu
• Yapay zeka tabanlı yanma rejimi sınıflandırması

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

References

1. 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
2. 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
3. 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
4. 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
5. 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
6. 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
7. 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.
8. 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

9. 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
10. 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.
11. Garcia, A. M., Le Bras, S., Prager, J., Häringer, M., & Polifke, W. (2022). Large eddy simulation of the dynamics of lean premixed flames using global reaction mechanisms calibrated for CH4–H2 fuel blends. Physics of Fluids, 34(9), Article 095105. https://doi.org/10.1063/5.0098898
12. George, N. B., Raghunathan, M., Unni, V. R., Sujith, R. I., Kurths, J., & Surovyatkina, E. (2022). Preventing a global transition to thermoacoustic instability by targeting local dynamics. Scientific Reports, 12, Article 9305. https://doi.org/10.1038/s41598-022-12951-6
13. Giannotta, A., Cherubini, S., & De Palma, P. (2023). The effect of hydrogen enrichment on thermoacoustic instabilities in laminar conical premixed methane/air flames. International Journal of Hydrogen Energy, 48(96), 37654–37665. https://doi.org/10.1016/j.ijhydene.2023.06.118
14. Habib, M. A., Abdulrahman, G. A. Q., Alquaity, A. B. S., & Qasem, N. A. A. (2024). Hydrogen combustion, production, and applications: A review. Alexandria Engineering Journal, 100, 182–207. https://doi.org/10.1016/j.aej.2024.05.030
15. Han, Z., Hossain, M. M., Wang, Y., Li, J., & Xu, C. (2020). Combustion stability monitoring through flame imaging and stacked sparse autoencoder based deep neural network. Applied Energy, 259, Article 114159. https://doi.org/10.1016/j.apenergy.2019.114159
16. Ji, L., Wang, J., Zhang, W., Wang, Y., Huang, Z., & Bai, X. S. (2024). Structure and thermoacoustic instability of turbulent swirling lean premixed methane/hydrogen/air flames in a model combustor. International Journal of Hydrogen Energy, 60, 890–901. https://doi.org/10.1016/j.ijhydene.2024.02.162
17. Kumar, A. D., Massey, J. C., Boxx, I., & Swaminathan, N. (2024). Effects of hydrogen enrichment on thermoacoustic and helical instabilities in swirl stabilised partially premixed flames. Flow, Turbulence and Combustion, 112(1), 689–727. https://doi.org/10.1007/s10494-023-00504-4
18. Lieuwen, T. C. (2021). Unsteady combustor physics. Cambridge University Press. https://doi.org/10.1017/9781108889001
19. Lieuwen, T. C., & Yang, V. (Eds.). (2005). Combustion instabilities in gas turbine engines: Operational experience, fundamental mechanisms, and modeling. American Institute of Aeronautics and Astronautics. https://doi.org/10.2514/5.9781600866807.0000.0000
20. Liu, C., Yu, L., Zhao, D., & Lu, X. (2026). Experimental studies on self-sustained combustion oscillation characteristics and flame/flow dynamics in a turbulent premixed annular combustor with different swirler configurations. Journal of Sound and Vibration 621, Article 119475. https://doi.org/10.1016/j.jsv.2025.119475
21. Lyu, Z., Fang, Y., Zhu, Z., Jia, X., Gao, X., & Wang, G. (2022). Prediction of acoustic pressure of the annular combustor using stacked long short-term memory network. Physics of Fluids, 34(5), Article 054109. https://doi.org/10.1063/5.0089146
22. Machado, D. A., Pinheiro, M. R., Villanueva, H. H. S., dos Santos Santana, P. H., & Krieger Filho, G. C. (2024). Investigating laminar burning velocity in ammonia-hydrogen mixtures using different kinetic mechanisms. In Proceedings of the 20th Brazilian Congress of Thermal Sciences and Engineering.
23. Mei, Y., Shuai, J., Zhou, N., Ren, W., & Ren, F. (2022). Flame propagation of premixed hydrogen-air explosions in bend pipes. Journal of Loss Prevention in the Process Industries, 77, Article 104790. https://doi.org/10.1016/j.jlp.2022.104790
24. Mohammadi, K., Immonen, J., Blackburn, L. D., Tuttle, J. F., Andersson, K., & Powell, K. M. (2023). A review on the application of machine learning for combustion in power generation applications. Reviews in Chemical Engineering, 39(6), 1027–1059. https://doi.org/10.1515/revce-2021-0107
25. Parente, A., & Swaminathan, N. (2024). Data-driven models and digital twins for sustainable combustion technologies. iScience, 27(4), Article 109349. https://doi.org/10.1016/j.isci.2024.109349
26. Poinsot, T., & Veynante, D. (2011). Theoretical and numerical combustion (3rd ed.). RT Edwards.
27. Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., Saisana, M., & Tarantola, S. (2008). Global sensitivity analysis: The primer. John Wiley & Sons.
28. Sudarsanan, S., Velamati, R. K., Alquaity, A. B. S., & Selvaraj, P. (2024). Impact of H₂ blending of methane on micro-diffusion combustion in a planar micro-combustor with splitter. Energies, 17(4), Article 970. https://doi.org/10.3390/en17040970
29. Tomlin, A. S., Fugger, C. A., & Caswell, A. W. (2020). Thermoacoustic instabilities in a three bluff body flow. In Proceedings of the American Institute of Aeronautics and Astronautics SciTech 2020 Forum.
30. Xu, B., Wang, Z., Zhou, H., Cao, W., Zhong, Z., Huang, W., & Nie, W. (2024). Detection of precursors of thermoacoustic instability in a swirled combustor using chaotic analysis and deep learning models. Aerospace, 11(6), Article 455. https://doi.org/10.3390/aerospace11060455
31. Zhang, S., Zhang, C., & Wang, B. (2024). CRK-PINN: A physics-informed neural network for solving combustion reaction kinetics ordinary differential equations. Combustion and Flame, 269, Article 113647. https://doi.org/10.1016/j.combustflame.2024.113647
32. Zhou, L., Song, Y., Ji, W., & Wei, H. (2022). Machine learning for combustion. Energy AI, 7, Article 100128. https://doi.org/10.1016/j.egyai.2021.100128