Browse Publications Technical Papers 2017-24-0040

Investigation of Sub-Grid Model Effect on the Accuracy of In-Cylinder LES of the TCC Engine under Motored Conditions 2017-24-0040

The increasing interest in the application of Large Eddy Simulation (LES) to Internal Combustion Engines (hereafter ICEs) flows is motivated by its capability to capture spatial and temporal evolution of turbulent flow structures. Furthermore, LES is universally recognized as capable of simulating highly unsteady and random phenomena driving cycle-to-cycle variability (CCV) and cycle-resolved events such as knock and misfire. Several quality criteria were proposed in the recent past to estimate LES uncertainty: however, definitive conclusions on LES quality criteria for ICEs are still far to be found.
This paper describes the application of LES quality criteria to the TCC-III single-cylinder optical engine from University of Michigan and GM Global R&D; the analyses are carried out under motored condition. In particular, attention is focused on sub-grid scale (SGS) model effects, which are evaluated using single grid estimators to compare three different sub-filter models: static Smagorinsky, dynamic Smagorinsky and dynamic structure model.
Information on LES quality criteria are cross-linked to the analysis of in-cylinder gas-dynamics and flow structures. These are in turn analyzed by comparing experimental results (Particle Image Velocimetry (PIV) velocity fields) with a dataset of consecutive LES cycles on four different cutting planes at engine-relevant crank angle positions. Finally, phase-dependent Proper Orthogonal Decomposition is used to draw further considerations on the connection between LES quality indices and the accuracy of simulation results and conclusions are drawn to be used as guidelines in future LES analyses of ICEs.


Subscribers can view annotate, and download all of SAE's content. Learn More »


Members save up to 18% off list price.
Login to see discount.
Special Offer: Download multiple Technical Papers each year? TechSelect is a cost-effective subscription option to select and download 12-100 full-text Technical Papers per year. Find more information here.