Effects of In-Cylinder Flow Simplifications on Turbulent Mixing at Varying Injection Timings in a Piston Bowl PPC Engine

In computational fluid dynamic simulations of partially premixed combustion engines it is common to find simplifications of the in cylinder flow conditions in order to save computational cost. One common simplification is to start the simulation at the moment of intake valve closing with an assumed initial flow condition, rather than making a full scavenging simulation. Another common simplification is the periodic sector assumption, limiting all sector cuts of the full cylinder to be identical periodic copies of each other. This work studies how such flow simplifications affect the spray injection and in turn the fuel/air mixing at different injection timings. Focus is put on the stratification of fuel concentration and gas temperature due to interaction of the spray, turbulence and piston geometry. The investigated engine setup consists of a light duty engine with a piston bowl and a five-hole injector. The simulations are performed under non-reacting conditions and utilize the large eddy simulation turbulence model. Both full cylinder mesh and sector mesh simulations are carried out to evaluate the effects of (a) turbulent vs non-turbulent initial conditions at intake valve closing, (b) sector periodicity assumption on the fuel/air mixing in the compression stroke under different PPC injection timings. The unresolved scales are modeled using a transported one-equation sub-grid scale (SGS) turbulence closure and the fuel spray is modelled using Lagrangian particle tracking (LPT). The simulations show that simple laminar initial conditions produce similar levels of mixing as advanced initial conditions with tilted swirl and anisotropic turbulence. The influence of sector periodicity restriction compared to a full cylinder simulation shows slightly increased levels of stratification in terms of temperature and fuel concentration. This difference is shown to depend on turbulent scales at the center of the cylinder inside the piston bowl and was strongest for the advanced injection timings but could also be observed for the earliest injections.