Yiqing Wang

Two-dimensional simulations incorporating detailed chemistry are conducted for detonation initiation induced by dual hot spots in a hydrogen/oxygen/argon mixture. The objective is to examine the transient behaviour of detonation initiation as facilitated by dual hot spots, and to elucidate the underlying mechanisms. Effects of hot spot pressure and distance on the detonation initiation process are assessed; and five typical initiation modes are identified. It is found that increasing the hot spot pressure promotes detonation initiation, but the impact of the distance between dual hot spots on detonation initiation is non-monotonic. During the initiation process, the initial hot spot autoignites, and forms the cylindrical shock waves. Then, the triple-shock structure, which is caused by wave collisions and consists of the longitudinal detonation wave, transverse detonation wave and cylindrical shock wave, dominates the detonation initiation behaviour. A simplified theoretical model is proposed to predict the triple-point path, whose curvature quantitatively indicates the diffraction intensity of transient detonation waves. The longitudinal detonation wave significantly diffracts when the curvature of the triple-point path is large, resulting in the failed detonation initiation. Conversely, when the curvature is small, slight diffraction effects fail to prevent the transient detonation wave from developing. The propagation of the transverse detonation wave is affected not only by the diffraction effects but also by the mixture reactivity. When the curvature of the triple-point trajectory is large, a strong cylindrical shock wave is required to compress the mixture, enhancing its reactivity to ensure the transverse detonation wave can propagate without decoupling.

Hydrogen-fueled internal combustion engines have been receiving increasing attention, particularly in applications such as on-road/off-road heavy-duty transport and combined heat and power. Direct injection (DI) of gaseous hydrogen (H₂) into the combustion chamber offers great potential for achieving high power density and engine efficiency, while mitigating the risk of backfire and reducing preignition. However, the numerical simulation of H₂ DI system remains a formidable challenge and there is a lack of well-established and validated practices for the modeling of H₂ DI in large-bore engines. Therefore, this study focuses on computational modeling of the mixture formation process in a heavy-duty optical engine where gaseous hydrogen is injected into the combustion chamber via a centrally located outward opening hollow-cone injector at 40 bar. Both Large Eddy Simulations (LES) and Reynolds Averaged Navier-Stokes (RANS) simulations are performed and evaluated by systematically comparing numerical predictions for H₂ distribution in different horizontal and vertical planes during the compression stroke against the planar laser-induced fluorescence measurements. Overall, the LES approach using Dynamic Structure model is found to have good predictive capabilities for the early jet penetration as well as the later H₂ distributions. However, the unsteady RANS approach with the Renormalization Group k-∈ model significantly underpredicts the H₂ mixing, even at similar mesh resolution to that used in LES. These results indicate that there is a need for the improvement of mixing sub-models within the RANS approach when applied to H₂ DI simulations.

Hydrogen has drawn great attention in recent years as a carbon-free fuel. The turbulent burning velocity (ST) is an important parameter for the design and modeling of hydrogen-fueled engines given the high propagation speed of hydrogen flames. It has been well documented that ST of hydrogen flames can be dramatically increased by thermo-diffusive effects which are sensitive to thermodynamic conditions. Previous studies have mainly focused on the pressure effect on ST of lean hydrogen flames, while the temperature effect has been largely ignored. In the present study, the turbulent burning velocity for a lean hydrogen/air mixture over a wide range of temperatures (300–641 K) and pressures (1–15 atm) is investigated through direct numerical simulations of statistically planar turbulent premixed flames. Results show that the variation of normalized turbulent burning velocity (ST/SL, where SL is the laminar flame speed) with temperature and pressure is mainly controlled by the variation of the stretching factor I0. While ST/SL is only marginally dependent on temperature at the atmospheric pressure, it exhibits a decreasing trend with temperature at an elevated pressure (10 atm). This is associated with different temperature dependencies of flame surface area enlargement at the two different pressures, despite the monotonically decreasing trends of I0 with temperature at both pressures. In addition, under engine-relevant conditions where the temperature and pressure increase simultaneously, the promotion effect of pressure is found to be largely canceled out by the suppression effect of temperature, leading to only a slight increase in I0 and ST/SL. The observed trends are further explained through detailed flame dynamic analysis. Furthermore, I0 at different temperatures and pressures is found to correlate very well with the enhancement of fuel consumption rate in the critically strained laminar flames. The present study elucidates the strong impact of temperature on ST of lean premixed hydrogen/air flames at elevated pressures and provides new insights into the modeling of ST, especially for engine-relevant conditions.