Enhancement of engine simulation using LES turbulence modeling and advanced numerical schemes

摘要: The goal of this study is to develop advanced numerical models and algorithms improve the accuracy engine spray combustion simulation. This developed a large eddy simulation (LES) turbulence model adaptive mesh refinement (AMR) enhance computational efficiency LES approach for modeling advantageous over traditional Reynolds Averaged Navier Stokes (RANS) due its capability obtain more detailed flow information by resolving large-scale structures which are strongly geometry dependent. current used one-equation, non-viscosity, dynamic structure sub-grid stress tensor also gradient method scalar fluxes. implementation was validated comparing predicted penetrations in non-evaporating diesel spray. present model, when coupled with breakup chemistry models, were able predict overall cylinder pressure history, heat release rate data, trends NOx soot emissions respect different injection timings EGR levels heavy-duty engine. Results indicated that could unsteadiness in-cylinder flows have potential provide compared RANS model. AMR transient It known inadequate spatial resolution can cause inaccuracy using stochastic Lagrangian particle over-estimated diffusion inappropriate