Implementation of the spectral line-based weighted-sum-of-gray-gases model in the finite volume method for radiation modeling in internal combustion engines

Abstract

It is well-known that the pollutant formation processes and temperature distribution in various combustion systems that operate at high temperatures are influenced by radiation heat transport. Detailed modeling of radiation transport in internal combustion (IC) engines demands additional computational power, and hence the calculation of radiation phenomenon is not commonly applied in IC engines. At the same time, current operating conditions in IC engines consider high temperatures and recirculation of exhaust gases that enhance gas radiation. Therefore, the application of radiation models is needed to increase the correctness of radiative absorption, combustion characteristics, and the formation of pollutant emissions. In this paper, the implementation and validation of the spectral line-based weighted-sum-of-gray-gases (SLW) model for calculating soot and gas radiation are performed. The SLW model is implemented in the computational fluid dynamics code AVL FIRE by programable user routines. The radiative transfer equation was calculated employing the finite volume method applicable for multiprocessing, moving meshes, and a mesh rezone procedure required for IC engine modeling. The validation of the SLW model is performed on one-dimensional geometric cases that include analytical results of radiation intensity, for which agreement within 10% of the relative error was achieved. Additionally, the SLW model is applied to compression ignition engine simulations, where the obtained results are compared with the measured pressure and concentrations of NO and soot emissions. The calculated heat losses through the wall boundary layer were around 12% of the total fuel energy, approximately 9.5% of the total fuel energy was lost due to the convective flow. 7%-8% of convection heat loss was due to the higher emission than absorption of participating CO2 and H2O gasses, and the rest are net soot losses. For the observed operating cases, the computational time is increased nearly double for SLW model than in the simulation without radiation. Finally, the results calculated using SLW indicate an improved agreement with the experimental mean pressure, temperature, soot, and NO concentrations compared to simulations without radiation.