Increasingly stringent pollutant and CO2 emission standards require engine manufacturers to investigate innovative solutions. Among these techniques, low-temperature combustion (LTC) concepts have a large potential to simultaneously reduce NOx emissions and fuel consumption. A promising manner to realize LTC consists of adopting ultra-lean mixtures, where the combustion evolution is controlled by a proper spatial distribution of fuels with different chemical reactivities. In this context, depending on the proportion and stratification of the fuels, the heat release can primarily depend on chemistry progression, leading to a Reactivity Controlled Compression Ignition (RCCI) mode, or on flame propagation, locally initiated by a high reactivity fuel. In this work, the combustion characteristics of a large-bore research engine are experimentally investigated. Natural gas is supplied into the intake port, while light fuel oil (LFO) is directly injected in the cylinder. An experimental campaign is carried out including sweeps of engine load, air/fuel proportion, LFO amount, valve timing, and intake air temperature. Global engine operating parameters as well as cylinder pressure traces are recorded and analyzed. Based on the available experimental data, a phenomenological model handling both chemistries of fuels with different reactivities and flame propagation is developed and validated. The model is based on a multi-zone approach, where auto-ignition chemistry is solved by a tabulated method to preserve the computational effort. The proposed numerical approach shows the ability to simulate the experimental data with good accuracy, using a fixed tuning constant set. A dedicated correlation is built to reproduce the expected in-cylinder distribution of the directly injected liquid fuel. Global performance and combustion parameters are predicted with an average error below 5%. The model demonstrates to correctly describe the behavior of the tested engine under different operating conditions and to capture the physics behind such advanced combustion concepts.
Development of a phenomenological model for the description of RCCI combustion in a dual-fuel marine internal combustion engine / De Bellis, V.; Malfi, E.; Lanotte, A.; Fasulo, G.; Bozza, F.; Cafari, A.; Caputo, G.; Hyvonen, J.. - In: APPLIED ENERGY. - ISSN 0306-2619. - 325:(2022), p. 119919. [10.1016/j.apenergy.2022.119919]
Development of a phenomenological model for the description of RCCI combustion in a dual-fuel marine internal combustion engine
De Bellis V.;Malfi E.;Lanotte A.;Fasulo G.;Bozza F.;
2022
Abstract
Increasingly stringent pollutant and CO2 emission standards require engine manufacturers to investigate innovative solutions. Among these techniques, low-temperature combustion (LTC) concepts have a large potential to simultaneously reduce NOx emissions and fuel consumption. A promising manner to realize LTC consists of adopting ultra-lean mixtures, where the combustion evolution is controlled by a proper spatial distribution of fuels with different chemical reactivities. In this context, depending on the proportion and stratification of the fuels, the heat release can primarily depend on chemistry progression, leading to a Reactivity Controlled Compression Ignition (RCCI) mode, or on flame propagation, locally initiated by a high reactivity fuel. In this work, the combustion characteristics of a large-bore research engine are experimentally investigated. Natural gas is supplied into the intake port, while light fuel oil (LFO) is directly injected in the cylinder. An experimental campaign is carried out including sweeps of engine load, air/fuel proportion, LFO amount, valve timing, and intake air temperature. Global engine operating parameters as well as cylinder pressure traces are recorded and analyzed. Based on the available experimental data, a phenomenological model handling both chemistries of fuels with different reactivities and flame propagation is developed and validated. The model is based on a multi-zone approach, where auto-ignition chemistry is solved by a tabulated method to preserve the computational effort. The proposed numerical approach shows the ability to simulate the experimental data with good accuracy, using a fixed tuning constant set. A dedicated correlation is built to reproduce the expected in-cylinder distribution of the directly injected liquid fuel. Global performance and combustion parameters are predicted with an average error below 5%. The model demonstrates to correctly describe the behavior of the tested engine under different operating conditions and to capture the physics behind such advanced combustion concepts.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
