Optimization of kinetic mechanism for hydrogen combustion based on machine learning

Abstract

The reduced mechanism based on the minimized reaction network method can effectively solve the rigidity problem in the numerical calculation of turbulent internal combustion engine. The optimization of dynamic parameters of the reduced mechanism is the key to reproduce the experimental data. In this work, the experimental data of ignition delay times and laminar flame speeds were taken as the optimization objectives based on the machine-learning model constructed by radial basis function interpolation method, and pre-exponential factors and activation energies of H₂ combustion mechanism were optimized. Compared with the origin mechanism, the performance of the optimized mechanism was significantly improved. The error of ignition delay times and laminar flame speeds was reduced by 24.3% and 26.8%, respectively, with 25% decrease in total mean error. The optimized mechanism was used to predict the ignition delay times, laminar flame speeds and species concentrations of jet stirred reactor, and the predicted results were in good agreement with experimental results. In addition, the differences of the key reactions of the combustion mechanism under specific working conditions were studied by sensitivity analysis. Therefore, the machine-learning model is a tool with broad application prospects to optimize various combustion mechanisms in a wide range of operating conditions.

Graphical abstract

Keywords

hydrogen combustion / machine learning / chemical kinetics / mechanism optimization

References

[1] Awad O I , Zhou B , Harrath K , Kadirgama K . Characteristics of NH₃/H₂ blend as carbon-free fuels: a review. International Journal of Hydrogen Energy, 2023, 48(96): 38077–38100

[2] LiXYaoXShentuJSunXLiJLiuMXuS. Combustion reaction mechanism construction by two-parameter rate constant method. Chemical Journal of Chinese Universities, 2020, 41(3): 512–520 (in Chinese)

[3] LiXShentuJLiYLiJWangJ. Combustion mechanism construction based on minimized reaction network: hydrogen-oxygen combustion. Chemical Journal of Chinese Universities, 2020, 41(4): 772–779 (in Chinese)

[4] Wang H , Sheen D A . Combustion kinetic model uncertainty quantification, propagation and minimization. Progress in Energy and Combustion Science, 2015, 47: 1–31

[5] Wang H , Yao M , Reitz R D . Development of a reduced primary reference fuel mechanism for internal combustion engine combustion simulations. Energy & Fuels, 2013, 27(12): 7843–7853

[6] Ra Y , Reitz R D . A reduced chemical kinetic model for IC engine combustion simulations with primary reference fuels. Combustion and Flame, 2008, 155(4): 713–738

[7] Lv D , Chen Y , Chen Y , Guo X , Chen H , Huang H . Development of a reduced diesel/PODE_n mechanism for diesel engine application. Energy Conversion and Management, 2019, 199: 112070

[8] Lin S , Sun W , Guo L , Cheng P , Sun Y , Zhang H . Development of a reduced mechanism of a three components surrogate fuel for the coal-to-liquid and diesel combustion simulation. Fuel, 2021, 294: 120370

[9] Lapene A , Debenest G , Quintard M , Castanier L M , Gerritsen M G , Kovscek A R . Kinetics oxidation of heavy oil. 2. Application of genetic algorithm for evaluation of kinetic parameters. Energy & Fuels, 2015, 29(2): 1119–1129

[10] Niu B , Jia M , Xu G , Chang Y , Xie M . Efficient approach for the optimization of skeletal chemical mechanisms with multiobjective genetic algorithm. Energy & Fuels, 2018, 32(6): 7086–7102

[11] Si J , Wang G , Li P , Mi J . Optimization of the global reaction mechanism for MILD combustion of methane using artificial neural network. Energy & Fuels, 2020, 34(3): 3805–3815

[12] Lin Q , Zheng J , Zou C , Cheng J , Li J , Xia W , Shi H . An improved 3-pentanone high temperature kinetic model using Bayesian optimization algorithm based on ignition delay times, flame speeds and species profiles. Fuel, 2020, 279: 118540

[13] Lin Q , Zou C , Liu S , Wang Y , Lu L , Peng C . An improved 2-pentanone low to high-temperature kinetic model using Bayesian optimization algorithm. Combustion and Flame, 2021, 231: 111453

[14] Li W , Zou C , Yao H , Lin Q , Fu R , Luo J . An optimized kinetic model for H₂/CO combustion in CO₂ diluent at elevated pressures. Combustion and Flame, 2022, 241: 112093

[15] Liu X , Wang Y , Bai Y , Yang W . Development of reduced and optimized mechanism for ammonia/hydrogen mixture based on genetic algorithm. Energy, 2023, 270: 126927

[16] Lin Q , Zou C , Luo J , Xia W , Li W , Peng C . A shock tube experiment and an improved high-temperature diisopropyl ketone model by Bayesian optimization. Combustion and Flame, 2022, 245: 112305

[17] Vollmer N I , Al R , Gernaey K V , Sin G . Synergistic optimization framework for the process synthesis and design of biorefineries. Frontiers of Chemical Science and Engineering, 2022, 16(2): 251–273

[18] Wang X , Li J , Zheng Y , Lin J . Smart systems engineering contributing to an intelligent carbon-neutral future: opportunities, challenges, and prospects. Frontiers of Chemical Science and Engineering, 2022, 16(6): 1023–1029

[19] Fang H , Zhou J , Wang Z , Qiu Z , Sun Y , Lin Y , Chen K , Zhou X , Pan M . Hybrid method integrating machine learning and particle swarm optimization for smart chemical process operations. Frontiers of Chemical Science and Engineering, 2022, 16(2): 274–287

[20] Chee E , Wong W C , Wang X . An integrated approach for machine-learning-based system identification of dynamical systems under control: application towards the model predictive control of a highly nonlinear reactor system. Frontiers of Chemical Science and Engineering, 2022, 16(2): 237–250

[21] Ludl P O , Heese R , Höller J , Asprion N , Bortz M . Using machine learning models to explore the solution space of large nonlinear systems underlying flowsheet simulations with constraints. Frontiers of Chemical Science and Engineering, 2022, 16(2): 183–197

[22] Kéromnès A , Metcalfe W K , Heufer K A , Donohoe N , Das A K , Sung C J , Herzler J , Naumann C , Griebel P , Mathieu O . . An experimental and detailed chemical kinetic modeling study of hydrogen and syngas mixture oxidation at elevated pressures. Combustion and Flame, 2013, 160(6): 995–1011

[23] Healy D , Kalitan D M , Aul C J , Petersen E L , Bourque G , Curran H J . Oxidation of C₁–C₅ alkane quinternary natural gas mixtures at high pressures. Energy & Fuels, 2010, 24(3): 1521–1528

[24] Olm C , Zsély I G , Varga T , Curran H J , Turányi T . Comparison of the performance of several recent syngas combustion mechanisms. Combustion and Flame, 2015, 162(5): 1793–1812

[25] Cantera: an object-oriented software toolkit for chemical kinetics, thermodynamics, and transport processes. Version 2.6.0. Pasadena: California Institute of Technology, 2022

[26] Sun W , Wang J , Huang C , Hansen N , Yang B . Providing effective constraints for developing ketene combustion mechanisms: a detailed kinetic investigation of diacetyl flames. Combustion and Flame, 2019, 205: 11–21

[27] Olm C , Zsély I G , Pálvölgyi R , Varga T , Nagy T , Curran H J , Turányi T . Comparison of the performance of several recent hydrogen combustion mechanisms. Combustion and Flame, 2014, 161(9): 2219–2234

[28] Pan L , Hu E , Deng F , Zhang Z , Huang Z . Effect of pressure and equivalence ratio on the ignition characteristics of dimethyl ether-hydrogen mixtures. International Journal of Hydrogen Energy, 2014, 39(33): 19212–19223

[29] Drakon A , Eremin A , Matveeva N , Mikheyeva E . The opposite influences of flame suppressants on the ignition of combustible mixtures behind shock waves. Combustion and Flame, 2017, 176: 592–598

[30] LVXHuEPengCMengXHuangZ. Measurements on laminar burning velocities of hydrogen/oxygen/diluents at elevated pressure and temperature. Journal of Aerospace Power, 2017, 32(7): 1599–1607 (in Chinese)

[31] Hu E , Huang Z , He J , Jin C , Zheng J . Experimental and numerical study on laminar burning characteristics of premixed methane-hydrogen-air flames. International Journal of Hydrogen Energy, 2009, 34(11): 4876–4888

[32] Qin X , Kobayashi H , Niioka T . Laminar burning velocity of hydrogen-air premixed flames at elevated pressure. Experimental Thermal and Fluid Science, 2000, 21(1): 58–63

[33] Huang Z , Zhang Y , Zeng K , Liu B , Wang Q , Jiang D . Measurements of laminar burning velocities for natural gas-hydrogen-air mixtures. Combustion and Flame, 2006, 146(1): 302–311

[34] Pang G A , Davidson D F , Hanson R K . Experimental study and modeling of shock tube ignition delay times for hydrogen-oxygen-argon mixtures at low temperatures. Proceedings of the Combustion Institute, 2009, 32(1): 181–188

[35] Zhou C W , Li Y , Burke U , Banyon C , Somers K P , Ding S , Khan S , Hargis J W , Sikes T , Mathieu O . . An experimental and chemical kinetic modeling study of 1,3-butadiene combustion: ignition delay time and laminar flame speed measurements. Combustion and Flame, 2018, 197: 423–438

[36] WangHYouXJoshiA VDavisS GLaskinAEgolfopoulosFLawC. USC Mech Version II. High-Temperature Combustion Reaction Model of H₂/CO/C₁/C₄ Compounds, 2007

[37] Behrooz H , Hayeri Y M . Machine learning applications in surface transportation systems: a literature review. Applied Sciences (Basel, Switzerland), 2022, 12(18): 9156

[38] Shakya A K , Pillai G , Chakrabarty S . Reinforcement learning algorithms: a brief survey. Expert Systems with Applications, 2023, 231: 120495

[39] Priore P , Ponte B , Puente J , Gómez A . Learning-based scheduling of flexible manufacturing systems using ensemble methods. Computers & Industrial Engineering, 2018, 126: 282–291

[40] Ryu J I , Kim K , Min K , Scarcelli R , Som S , Kim K S , Temme J E , Kweon C B M , Lee T . Data-driven chemical kinetic reaction mechanism for F-24 jet fuel ignition. Fuel, 2021, 290: 119508

[41] Nelder J A , Mead R . A simplex method for function minimization. Computer Journal, 1964, 7(4): 308–313

[42] Sutherland J W , Michael J V , Pirraglia A N , Nesbitt F L , Klemm R B . Rate constant for the reaction of O(3P) with H₂ by the flash photolysis-shock tube and flash photolysis-resonance fluorescence techniques; 504 K ≤ T ≤ 2495 K. Symposium (International) on Combustion, 1988, 21(1): 929–941

[43] Ryu S O , Hwang S M , Rabinowitz M J . Rate coefficient of the O + H₂ = OH + H reaction determined via shock tube-laser absorption spectroscopy. Chemical Physics Letters, 1995, 242(3): 279–284

[44] Mousavipour S H , Saheb V . Theoretical study on the kinetic and mechanism of H + HO₂ reaction. Bulletin of the Chemical Society of Japan, 2007, 80(10): 1901–1913

[45] Yang H , Gardiner W C , Shin K S , Fujii N . Shock tube study of the rate coefficient of H + O₂ → OH + O. Chemical Physics Letters, 1994, 231(4): 449–453

[46] Du H , Hessler J P . Rate coefficient for the reaction H + O₂ → OH + O: results at high temperatures, 2000 to 5300 K. Journal of Chemical Physics, 1992, 96(2): 1077–1092

[47] Shin K S , Michael J V . Rate constants for the reactions H + O₂ → OH + O and D + O₂ → OD + O over the temperature range 1085–2278 K by the laser photolysis-shock tube technique. Journal of Chemical Physics, 1991, 95(1): 262–273

[48] Pirraglia A N , Michael J V , Sutherland J W , Klemm R B . A flash photolysis-shock tube kinetic study of the hydrogen atom reaction with oxygen: H + O₂ → OH + O (962 K ≤ T ≤ ltoreq. 1705 K) and H + O₂ + Ar → HO₂ + Ar (746 K ≤ T ≤ 987 K). Journal of Physical Chemistry, 1989, 93(1): 282–291

[49] Le Cong T , Dagaut P . Experimental and detailed modeling study of the effect of water vapor on the kinetics of combustion of hydrogen and natural gas, impact on NO_x. Energy & Fuels, 2009, 23(2): 725–734

Competing interests

The authors declare that they have no competing interests.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant No. U20A20151) and the National Key R&D Program of China (Grant No. 2018YFA0702400).

Electronic Supplementary Material

Supplementary material is available in the online version of this article at https://doi.org/10.1007/s11705-024-2487-0 and is accessible for authorized users.

RIGHTS & PERMISSIONS

2024 Higher Education Press