Development of Chemistry-Based Laminar Flame Speed Correlation for Part-Load SI Conditions and Validation in a GDI Research Engine

Gasoline-ethanol blend formulation to mimic laminar flame speed and auto-ignition quality in automotive engines

“…1D chemistry-based simulations are in reasonable agreement with the observations by Syed at al. [36]. In Fig.…”

“…Such surrogates are formulated with a methodology developed by the authors and, to assess their validity, the combustion-relevant properties are compared with a wide literature dataset. In the following paragraph, a recently proposed methodology [35,36] to formulate LFS correlations is briefly recalled and results from chemistry-based 1D premixed flame simulations are presented for five selected engine-relevant blends (E0, E5, E10, E20, and E85). The aforementioned methodology is used to derive correlations, providing an accurate estimation of LFS for gasoline-ethanol blends at typical full-load conditions of GDI engines, to be implemented in 3D-CFD codes.…”

“…In addition, Jerzembeck et al [45] and Sileghem et al [46] demonstrated that a TRF fuel surrogate approach is suitable to retain a good accuracy also for lean and rich mixtures. In previous works by the authors [35,36], two different fuel surrogates, to be used with different chemical kinetics mechanisms, were chosen to compute LFS of a typical European gasoline. As reported in [47], refined fuels are complex mixtures of hydrocarbons that often vary based on both geographic location and season.…”

“…The difference in terms of stoichiometric air-to-fuel ratio (-1.12%) is acceptable as no significant impact in the LFS prediction is expected. In order to further assess the capability of the methodology to generate a surrogate able to represent the premixed flame propagating characteristics of a given fuel, 1D calculations were carried out using the methodology proposed by the authors in [36]. Both the CRECK-MECH chemical kinetics mechanism [56] and the Andrae's one [43] were used along with the proposed TRF surrogate, in order to minimize the error related to the use of a specific mechanism beyond its validation range, as discussed in [35,36].…”

“…For each of the selected ECs, 1D LFS calculations were carried out over a wide range of equivalence ratios (in order to account for the mixture inhomogeneity in the combustion chamber due to direct-injection) and dilution rates: ranges from 0.4 to 2.0, while EGR mass fraction from 0 to 20. A numerical methodology, extensively discussed by the authors in [35,36], was used to fit the s L dataset calculated at the ECs. In the present study, the latest version of the methodology [37] is used to derive the s L correlations.…”