Laminar flame speeds under engine-relevant conditions: Uncertainty quantification and minimization in spherically expanding flame experiments

The spherically expanding flame method is the only approach for measuring laminar flame speeds at thermodynamic states that are relevant to engines. In the present study, a comprehensive evaluation of data obtained under constant pressure and constant volume conditions was carried out through experiments, development of a mathematically rigorous method for uncertainty quantification and propagation, and advancement of numerical models that describe the experiments accurately. The proposed uncertainty characterization approach accounts for parameters related to all measurements, data processing, and finally data interpretation. With the aid of direct numerical simulations, an alternative approach was proposed to derive laminar flame speeds in constant pressure experiments by eliminating the need for using extrapolation equations developed based on simplifying assumptions, which are known to be susceptible to major errors under certain conditions. The propagation of spherical flames under constant volume conditions was investigated through experiments carried out in an entirely spherical chamber and the use of two numerical models. The first involves the solution of the one-dimensional conservation equations of mass, species, and energy while accounting for pressure rise. The second model was developed based on thermodynamics similarly to existing literature, but radiation loss was introduced at the optically thin limit and approximations were made to allow for re-absorption with minimum computational cost. It was shown that neglecting radiation in constant volume experiments could introduce errors as high as 15%. Incorporating the aforementioned techniques, laminar flame speeds were measured and reported with properly quantified uncertainties for flames of synthesis gas for pressures ranging from 3 to 30 atm, and unburned mixture temperatures ranging from 298 to 550 K. Selected measurements were carried out as well for methane and propane flames for pressures ranging from 3 to 7 atm, and unburned mixture temperature of 298 K. The approaches introduced in this study allow for the determination of laminar flame speeds with notably reduced uncertainties under conditions of relevance to engines, which has major implications for the validation of kinetic models of surrogate and real fuels.