Hydrogen-fuelled internal combustion engines (H2ICEs) can reduce the CO2 emissions of the transport sector. In literature, several 3D Computational Fluid Dynamics (CFD) methodologies to simulate H2ICE combustion are used. However, the wide range of conditions under which H2ICEs operate results in a large variety of numerical setups. In this work, Three-Zones Extended Coherent Flamelet Model (ECFM-3Z) and Complex Chemistry with Turbulent Flame Speed Closure model (CC-TFC) are analysed, and a robust methodology is proposed. Flame speed analyses on a combustion plane show that with low turbulence, CC-TFC flame speed tends to the correspondent laminar value, while the ECFM-3Z one tends to zero. As turbulence increases, the ECFM-3Z flame speed becomes higher than CC-TFC one. This behaviour is confirmed by in-cylinder simulations performed on a port fuel injection (PFI) lean H2ICE and on a stoichiometric hydrogen direct injection (HDI) engine, operated at part and full load. With ECFM-3Z, the MFB10-50 is 22 % shorter than experiments with low turbulence and 56 % faster with high turbulence, requiring significant calibration. As for CC-TFC, with a single setup, the MFB10-50 is less than 3 % longer than experiments in both PFI engine conditions and 15 % faster in the HDI engine at part load, with a minor tuning needed only for the full load condition. The different models behaviour with low turbulence affects also the wall-flame interaction, as with ECFM-3Z the flame is artificially quenched when approaching the walls, which leads to a 30 % less heat transferred to the walls compared to CC-TFC.
Predictive capabilities of flamelet combustion models in hydrogen engines across a broad range of operating conditions / Madia, Manuel; Breda, Sebastiano; Berni, Fabio; Cicalese, Giuseppe; Fontanesi, Stefano. - In: FUEL. - ISSN 0016-2361. - 405:(2026), pp. 1-20. [10.1016/j.fuel.2025.136580]
Predictive capabilities of flamelet combustion models in hydrogen engines across a broad range of operating conditions
Madia, Manuel
;Breda, Sebastiano;Berni, Fabio;Cicalese, Giuseppe;Fontanesi, Stefano
2026
Abstract
Hydrogen-fuelled internal combustion engines (H2ICEs) can reduce the CO2 emissions of the transport sector. In literature, several 3D Computational Fluid Dynamics (CFD) methodologies to simulate H2ICE combustion are used. However, the wide range of conditions under which H2ICEs operate results in a large variety of numerical setups. In this work, Three-Zones Extended Coherent Flamelet Model (ECFM-3Z) and Complex Chemistry with Turbulent Flame Speed Closure model (CC-TFC) are analysed, and a robust methodology is proposed. Flame speed analyses on a combustion plane show that with low turbulence, CC-TFC flame speed tends to the correspondent laminar value, while the ECFM-3Z one tends to zero. As turbulence increases, the ECFM-3Z flame speed becomes higher than CC-TFC one. This behaviour is confirmed by in-cylinder simulations performed on a port fuel injection (PFI) lean H2ICE and on a stoichiometric hydrogen direct injection (HDI) engine, operated at part and full load. With ECFM-3Z, the MFB10-50 is 22 % shorter than experiments with low turbulence and 56 % faster with high turbulence, requiring significant calibration. As for CC-TFC, with a single setup, the MFB10-50 is less than 3 % longer than experiments in both PFI engine conditions and 15 % faster in the HDI engine at part load, with a minor tuning needed only for the full load condition. The different models behaviour with low turbulence affects also the wall-flame interaction, as with ECFM-3Z the flame is artificially quenched when approaching the walls, which leads to a 30 % less heat transferred to the walls compared to CC-TFC.| File | Dimensione | Formato | |
|---|---|---|---|
|
1-s2.0-S0016236125023051-main.pdf
Open access
Tipologia:
VOR - Versione pubblicata dall'editore
Licenza:
[IR] creative-commons
Dimensione
7.58 MB
Formato
Adobe PDF
|
7.58 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate

I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris




