In the new combustion strategies such as RCCI and dual-fuel combustion, the diesel pilot injection plays a pivotal role as it determines the ignition characteristics of the mixture and ultimately the combustion and emission performance. In this regard, equivalence ratio distribution resulted from the pilot injection becomes very important. In this work, computation study is carried out using KIVA-3V to simulate the engine compression stroke from intake valve close (IVC) to close to TDC so as to investigate the impact of piston geometry, injection start timing and flow initialization on the equivalence ratio distribution from a pilot injection in HSDI engine. Two piston geometry (the stock piston with bowl-in piston shape and a flat piston), three injection timing (80 CA, 60 CA and 40 CA BTDC) and three velocity initializations (Bessel fit with constant value of 3.11 and 2.2; and a velocity field obtained from PIV measurement on a steady-state bench flow test facility with minimum valve lift) were considered. It is found that the low swirl ratio generated by the flat piston caused a fuel rich core formed in the central region of the cylinder while most of the fuel vapor was trapped in the re-entrant bowl region for the flat piston. Injection start timing is the dominant factor affecting the homogeneity of the mixture as advanced injection provided more uniform mixture towards TDC. The cons of the early injection though, is a small amount of fuel spray will penetrate into the squish region even with the small injection spread angle, which may lead to UHC and CO emissions. Flow field measured by PIV suggest there could be a wake region at the IVC timing. Initializing the velocity field using the experimental data resulted in drastic difference among individual plumes during and after the injection especially for the bowl-in-piston geometry, emphasizing the importance of full geometry and more accurate flow initialization in the 3D engine simulations. In comparison, the flow initialization with different Bessel fit constant did not show remarkable difference although an α value of 3.11 provided more homogeneous mixture in the central region.
ASJC Scopus subject areas
- Automotive Engineering
- Safety, Risk, Reliability and Quality
- Industrial and Manufacturing Engineering