Predicting Hydrogen Combustion Characteristics Under Nitrogen Dilution: A Comparative Evaluation of GPR, MLP, and DNN Models

Asadi, Ali; Noferesti, Morteza

doi:10.22044/jadm.2026.16390.2761

Document Type : Original/Review Paper

Authors

Ali Asadi ¹
Morteza Noferesti ²

¹ Department of Mechanical Engineering, Bozorgmehr University of Qaenat, Qaen, Iran.

² Department of Computer Engineering, Bozorgmehr University of Qaenat, Qaen, Iran.

https://doi.org/10.22044/jadm.2026.16390.2761

Abstract

Hydrogen combustion has emerged as a pivotal technology for decarbonizing the energy sector, offering a clean and sustainable alternative to fossil fuels. This study investigates hydrogen combustion dynamics in a perfectly stirred reactor (PSR) under steady-state, non-premixed conditions. It is employing CHEMKIN-based simulations to analyze the effects of nitrogen dilution, operating pressure, and equivalence ratio on flame temperature and NOx emissions. The parametric results reveal that nitrogen dilution reduces flame temperature by up to 28% and suppresses NOx emissions by 15–40%, while elevated pressure promotes higher flame temperatures and increased NOx formation. Peak temperature and NOx concentrations are observed under stoichiometric conditions (φ = 1.0), with both quantities decreasing under lean and rich mixture conditions. To enable rapid and accurate prediction of these combustion characteristics, three machine learning models were developed and benchmarked on the CHEMKIN-generated dataset: Gaussian Process Regression, Multilayer Perceptron, and Deep Neural Network. GPR demonstrated the best overall predictive performance, achieving the lowest MAE for temperature prediction (MAE = 1.77) and major species concentrations. Although the DNN produced competitive results (MAE = 58.28), it demanded approximately five times more computational resources than GPR, without a proportional gain in predictive accuracy. All three models, maintained prediction errors below 20% across the investigated parameter space, confirming their viability as efficient tools for hydrogen combustion optimization. These findings demonstrate that physics-informed machine learning models, when combined with high-fidelity combustion simulations, offer a powerful and computationally efficient pathway toward accelerating the design and optimization of clean hydrogen energy systems.

Keywords

Main Subjects

H.6.2.4. Neural nets

References

[1] A. Mirzabaev and Q. Chen, "Chapter 18: Sustainable development goal 7: Affordable and clean energy," in Handbook on Public Policy and Food Security. Cheltenham, UK: Edward Elgar Publishing, 2024, pp. 185–193.

[2] S. Goke, S. Terhaar, S. Schimek, K. Göckeler, and C. O. Paschereit, "Combustion of natural gas, hydrogen and bio-fuels at ultra-wet conditions," in Proc. ASME Turbo Expo, vol. 2: Combustion, Fuels and Emissions, Parts A and B, 2011, pp. 659–670.

[3] S. Göke, M. Füri, G. Bourque, B. Bobusch, K. Göckeler, O. Krüger, S. Schimek, S. Terhaar, and C. O. Paschereit, "Influence of steam dilution on the combustion of natural gas and hydrogen in premixed and rich-quench-lean combustors," Fuel Processing Technology, vol. 107, pp. 14–22, 2013.

[4] S. Dybe, F. Güthe, M. Bartlett, P. Stathopoulos, and C. O. Paschereit, "Experimental characterization of the combustion in fuel flexible humid power cycles," in Proc. ASME Turbo Expo, vol. 3A: Combustion, Fuels, and Emissions, 2021, pp. 03–04012.

[5] Y. Lyu, P. Qiu, L. Liu, C. Yang, and S. Sun, "Effects of steam dilution on laminar flame speeds of H2/air/H2O mixtures at atmospheric and elevated pressures," International Journal of Hydrogen Energy, vol. 43, no. 15, pp. 7538–7549, 2018.

[6] K. Zhang, Y. Shen, and C. Duwig, "Finite rate simulations and analyses of wet/distributed flame structure in swirl-stabilized combustion," Fuel, vol. 289, art. 119922, 2021.

[7] O. Krüger, S. Terhaar, C. O. Paschereit, and C. Duwig, "Large eddy simulations of hydrogen oxidation at ultra-wet conditions in a model gas turbine combustor applying detailed chemistry," Journal of Engineering for Gas Turbines and Power, vol. 135, no. 2, art. 021501, 2013.

[8] R. Palulli, S. Dybe, K. Zhang, F. Güthe, P. R. Alemela, C. O. Paschereit, and C. Duwig, "Characterisation of non-premixed, swirl-stabilised, wet hydrogen/air flame using large eddy simulation," Fuel, vol. 350, art. 128710, 2023.

[9] G. J. Rortveit, J. E. Hustad, S.-C. Li, and F. A. Williams, "Effects of diluents on NOx formation in hydrogen counterflow flames," Combustion and Flame, vol. 130, no. 1, pp. 48–61, 2002.

[10] M. Skottene and K. E. Rian, "A study of NOx formation in hydrogen flames," International Journal of Hydrogen Energy, vol. 32, no. 15, pp. 3572–3585, 2007.

[11] J. W. Bozzelli and A. M. Dean, "O + NNH: A possible new route for NOx formation in flames," International Journal of Chemical Kinetics, vol. 27, no. 11, pp. 1097–1109, 1995.

[12] B. Breer, H. Rajagopalan, C. Godbold, H. Johnson, B. Emerson, V. Acharya, W. Sun, D. Noble, and T. Lieuwen, "Numerical investigation of NOx production from premixed hydrogen/methane fuel blends," Combustion and Flame, vol. 255, art. 112920, 2023.

[13] S. Taamallah, K. Vogiatzaki, F. M. Alzahrani, E. M. A. Mokheimer, M. A. Habib, and A. F. Ghoniem, "Fuel flexibility, stability and emissions in premixed hydrogen-rich gas turbine combustion: Technology, fundamentals, and numerical simulations," Applied Energy, vol. 154, pp. 1020–1047, 2015.

[14] I. Chterev and I. Boxx, "Effect of hydrogen enrichment on the dynamics of a lean technically premixed elevated pressure flame," Combustion and Flame, vol. 225, pp. 149–159, 2021.

[15] N. Tathawadekar, N. A. K. Doan, C. F. Silva, and N. Thuerey, "Modeling of the nonlinear flame response of a Bunsen-type flame via multi-layer perceptron," Proceedings of the Combustion Institute, vol. 38, no. 4, pp. 6261–6269, 2021.

[16] V. Yadav, M. Casel, and A. Ghani, "Physics-informed recurrent neural networks for linear and nonlinear flame dynamics," Proceedings of the Combustion Institute, vol. 39, no. 2, pp. 1597–1606, 2023.

[17] M. McCartney, M. Haeringer, and W. Polifke, "Comparison of machine learning algorithms in the interpolation and extrapolation of flame describing functions," Journal of Engineering for Gas Turbines and Power, vol. 142, no. 6, art. 061009, 2020.

[18] Y. Shen and A. S. Morgans, "Predicting the effect of hydrogen enrichment on the flame describing function using machine learning," International Journal of Hydrogen Energy, vol. 79, pp. 267–276, 2024.

[19] G. Mao, T. Shi, C. Mao, and P. Wang, "Prediction of NOx emission from two-stage combustion of NH3–H2 mixtures under various conditions using artificial neural networks," International Journal of Hydrogen Energy, vol. 49, pp. 1414–1424, 2024.

[20] K. Ramalingam, S. Vellaiyan, M. Kandasamy, D. Chandran, and R. Raviadaran, "An experimental study and ANN analysis of utilizing ammonia as a hydrogen carrier by real-time emulsion fuel injection to promote cleaner combustion," Results in Engineering, vol. 21, art. 101946, 2024.

[21] K. J. Reddy, G. A. Prasad Rao, R. M. Reddy, and Ağbulut, "Artificial intelligence based forecasting of dual-fuel mode CI engine behaviors powered with the hydrogen-diesel blends," International Journal of Hydrogen Energy, vol. 87, pp. 1074–1086, 2024.

[22] U. Rajak, Ağbulut, A. Dasore, and T. N. Verma, "Artificial intelligence based prediction of energy efficiency and tailpipe emissions of soybean methyl ester fuelled CI engine under variable compression ratios," Energy, vol. 294, art. 130861, 2024.

[23] N. J. I. Banta, N. Patrick, F. Offole, and R. Mouangue, "Machine learning models for the prediction of turbulent combustion speed for hydrogen-natural gas spark ignition engines," Heliyon, vol. 10, no. 9, 2024.

[24] S. R. Turns, An Introduction to Combustion: Concepts and Applications. New York: McGraw-Hill, 2000.

[25] E.Moradi, "An MLP-based deep neural network incorpoating SMOTE-Tomek approach for robust prediction of liver disorders," Journal of AI and Data Mining, vol. 13, no. 4, pp. 467–479, 2025.

[26] E. Yadolahi and S. Abolmaali, "Improving the efficiency of semantic segmentation implemented in spiking neural networks," Journal of AI and Data Mining, vol. 13, no. 1, pp. 25–39, 2025.

[27] C. B. Oh, E. J. Lee, and G. J. Jung, "Unsteady auto-ignition of hydrogen in a perfectly stirred reactor with oscillating residence times," Chemical Engineering Science, vol. 66, no. 20, pp. 4605–4614, 2011.

Predicting Hydrogen Combustion Characteristics Under Nitrogen Dilution: A Comparative Evaluation of GPR, MLP, and DNN Models

References

References

Volume 14, Issue 3July 2026Pages 275-289

Volume 14, Issue 3
July 2026
Pages 275-289