Assessment of Wall Heat Transfer Models for Premixed-Charge Engine Combustion Computations

Two-dimensional computations of premixed-charge engine combustion were made using the KIVA-II code. The purpose of the study was to assess the influence of heat transfer and turbulence model boundary conditions on engine combustion predictions. Combustion was modeled using a laminar- and turbulent-characteristic-time model. Flow through the piston-cylinder-ring crevice was accounted for using a phenomenological crevice-flow model. The predictions were compared to existing cylinder pressure and wall heat transfer experimental data under motoring and fired conditions, at two engine speeds. Two different wall heat transfer model formulations were considered. The first is the standard wall function method. The second is based on solutions to the one-dimensional unsteady energy equation, formulated such that the standard wall function method is recovered in the quasi-steady limit. Turbulence was modeled using the standard k-ε turbulence model equations. However, the turbulence model boundary conditions were modified to account for compressibility effects by using a coordinate transformation in the wall region. The results show that the details of wall heat transfer and turbulence model boundary conditions influence heat transfer predictions greatly through their influence on the flame speed and the flame structure in the vicinity of the wall. Inclusion of compressibility and unsteadiness effects leads to increased wall heat flux values that agree better with measurements.