The High-Fidelity Industrial LES/DNS (HiFiLeD) Symposium is taking place, 14th - 16th November in Brussels.
The aim of the symposium is to examine new areas of turbulence research and the ability to perform reliable high-fidelity large-eddy simulations (LES) and direct numerical simulations (DNS) for industrial relevant flow configurations. This has attracted extensive research in recent years, stimulated by other emerging areas such as Big Date, Artificial Intelligence (AI) and Machine Learning.
These topic areas will be the main focus of the HiFiLeD Symposium, exploring all aspects relating to each one. From issues concerning the complexity, reliability, accuracy and uncertainties in generating the HiFiLeD data, to their application towards turbulence and transition modelling.
The Centre for Modelling & Simulation (CFMS) is delighted to announce that our CFD Engineer, Andrei Cimpoeru will be presenting at the symposium (please see abstract below). If you would like to make an appointment to speak with CFMS during HiFiLeD, call 0117 906 1100 or email firstname.lastname@example.org.
For more information about the symposium, click here.
An Industrial High Order Flux Reconstruction Framework To Tackle Turbulent Flows
Andrei Cimpoeru, The Centre for Modelling & Simulation, Bristol, United Kingdom
Mark Allan, David Standingford, Zenotech Ltd, Bristol, United Kingdom
Most of the current industrial CFD tools rely on methods for which spatial order of accuracy is less than three. At the present time, these methods are used in conjunction with Reynolds Averaged Navier-Stokes (RANS) equations to provide design guidance solutions on unstructured meshes running on typical High Performance Computing (HPC) clusters within a few hours. These methods are robust and efficient but they are lacking in predicting massively separated vortical flows across most of the industrial sectors: Aerospace (shock induced separations, flows over cavities, high-lift scenarios), Automotive (vortical flows) and Renewable (wind farm wake interactions). The reason for this is that the current industrial methods are too dissipative. This could be an issue when investigating aero-acoustic problems with Large Eddy Simulation (LES). Recent techniques in numerical modelling have proven High Order Methods (HOMS) to be successful in tackling some of the challenges mentioned. They offer the advantage of high accuracy on complex unstructured meshes, are easy to parallelise and offer portability to modern platforms such as GPUs. This makes HOMS a good candidate for the next generation of CFD solvers. In the present work, zCFD, an industrial High Order framework based on the Flux Reconstruction method developed in  is discussed. Results from the UK Aerospace Technology Institute funded project Hyperflux++ are presented based on two test cases. The first refers to the near field acoustic predictions for transonic cavity flows using Detached Eddy Simulation (DES). Fig. 1 shows the turbulent flow field obtained using the fourth order (P3) flux reconstruction method with zCFD. The flow was computed using DES  and compared against wind tunnel data.