Prediction of Soot Mass and Particle Size in a High-boosted Diesel Engine Using Large Eddy Simulation

In the present study, a mixed timescale subgrid model of large eddy simulation (LES) is applied to simulate detailed mixture formation, combustion and soot formation influenced by turbulence in diesel engine combustion. This model is able to account for near wall turbulence without an explicit damping function. Therefore, the boundary conditions can be provided accurately. The combustion model uses a direct integration approach with an explicit Ordinary Differential Equation (ODE) solver called ERENA, and additionally parallelized by OpenMP. The Diesel oil surrogate mechanism was used which was developed at Chalmers University of Technology, consisting of 74 species and 320 reactions. The soot mass production within a computation cell is determined from a phenomenological soot formation model developed by WASEDA University. The model is combined with the LES code mentioned above, including the following important steps: particle inception in which naphthalene grows irreversibly to form soot, surface growth with the addition of C2H2, surface oxidation with OH radical and O2 attack, particle coagulation and particle agglomeration. Computational results are compared with experimental data from a high-boosted heavy-duty diesel engine. The predicted soot emissions are compared with experimental data under various EGR conditions. The results show that the in-cylinder pressure and the heat release rate obtained from the engine test are in good agreement with calculation data. In the soot emission calculation, the simulated results showed the exponential increases with increasing EGR rate. Furthermore, the steep rise of soot mass increase with increasing EGR rate from 30% EGR was reproduced. The soot mass and particle size characteristics against the EGR rate were analyzed. The process and spatial distribution of soot formation in a high-boosted heavy duty engine were studied.