Turbulence approaches for numerical predictions of vehicle-like afterbody vortex flows

Abstract

The numerical prediction of aerodynamic characteristics for vehicles is crucial to both industry and academia, with various numerical approaches playing a critical role in accurately resolving flow fields. This study aims to evaluate the effectiveness of three typical numerical approaches, including RANS, IDDES, and LES in predicting the afterbody vortex flows of a generic model, specifically a slanted-base cylinder. This study involved analyzing aerodynamic coefficients, time-averaged surface flow, time-averaged surrounding flow and transient flow, revealing the capabilities of each approach. RANS offers acceptable accuracy in predicting time-averaged aerodynamic coefficients and surface flow patterns, though it falls short in capturing time-varying physical quantities. LES, despite its higher computational cost, provides a more accurate prediction for both time-averaged and transient flow behaviors, particularly in capturing flow instabilities and multi-scale fluctuations. IDDES can be prioritized when a rough understanding of transient characteristics is sufficient. This study highlights the unique strengths and limitations of three typical numerical approaches in predicting vehicle-like afterbody vortex flows, guiding the selection of appropriate methods based on specific research needs.