Abstract
The particle emission is still a significant problem of internal combustion engines. This pollutant is suspected to increase the probability of lung cancer occurrence. Particle filters or alternative (oxygen containing) fuels provide an improvement but the careful diffusion flame optimization can bring the efficient and comparatively cheap solution, as well.
The authors have developed the model of a combustion in a diesel turbocharged engine focused to soot emission simulation. Both zero-dimensional (0-D) and a three-dimensional (3-D) concepts have been used. The latter one is based on the platform of a 3-D code for engine thermodynamic simulation KIVA 3.
The 3-D numerical simulation gives a detailed preview of chemical processes in a combustion chamber. Optimization of combustion chamber geometry and a fuel injection equipment may be based on it. The authors used originally a Hiroyasu soot model in combination with the Reitz Characteristic Time Combustion Model (CTCM). This combination proved to be rather inadequate for the case of a truck engine comparing the predictions with experiments. The too high sensitivity of a predicted soot emission to a residual unburned fuel was found. The results were compared with measured data available at the ETH using the original methodology of engine endoscopy. No reasonably realistic change of model parameters was able to correct these results. Therefore, a new model model concerning the rate of soot formation and oxidation has been developed and CTCM has been slightly adapted by a correction of turbulent mixing time. The proposed model was preliminary tested in the 0-D case. After it, the preliminary calibrated model was implemented into KIVA 3, comparing it with Hiroyasu soot model and with experimental results. At this level of development, new model has proven to be more prospective for predictions of soot formation. Nevertheless, it should be tested for more cases of diesel engines in the future.