LES-RANS coupling for the simulation of complex flows

用于模拟复杂流动的 LES-RANS 耦合

• 批准号: 5405882
• 负责人: Professor Dr.-Ing. Jochen Fröhlich
• 金额: --
• 依托单位: Institut für Strömungsmechanik
• 依托单位国家: 德国
• 项目类别: Research Units
• 财政年份: 2003
• 资助国家: 德国
• 起止时间: 2002-12-31 至 2009-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/5405882?language=en
• 关键词: LES; RANS; coupling; simulation; complex

LES has clear superiority over RANS-methods for complex flows, especially when large-scale structures dominate the turbulent transport, in transitional situations and when dynamic forces and noise generation are calculated. However, due to the extremely high resolution and time-averaging requirements, LES still is too costly to be applied directly to the flows of practical interest. On the other hand, RANS can determine the mean flow with engineering accuracy at much lower cost in many cases, especially attached flows, which often prevail in subareas of complex flows. Hence, a hybrid method using RANS in subareas with simpler flow behaviour and LES only in the more critical areas appears to be the ideal approach for complex high Reynolds number flows. The present project aims to develop and test such methods in a so-called zonal approach: RANS and LES will be applied in pre-defined subareas of the calculation domain and linked by suitable coupling conditions. Three coupling situations are considered: the LES-region is upstream of the RANS-region, RANS is applied upstream of LES, and the interface is tangential to the flow. The French and German partner will jointly develop coupling procedures for all three types and test them first for simple channel flow. The methods are then applied for testing and further improvement to increasingly complex flows taken from the common test cases of the group, such as to flow over hills, in a combustor and over a car-shaped body by the German partner and to separating boundary layers and finally the flow over a complete airfoil by the French partner.

对于复杂流动，特别是当大尺度结构主导湍流输运时，在过渡情况下以及计算动力和噪声产生时，LES比ranss方法具有明显的优势。然而，由于极高的分辨率和时间平均要求，LES仍然过于昂贵，无法直接应用于实际兴趣流。另一方面，在许多情况下，RANS可以以较低的成本以工程精度确定平均流量，特别是在复杂流的子区域中经常出现的附加流。因此，在流动行为较简单的子区域使用RANS，而仅在更关键的区域使用LES的混合方法似乎是复杂的高雷诺数流动的理想方法。本项目旨在以所谓的区域办法发展和试验这种方法：区域系统和区域系统将应用于计算领域的预先确定的子领域，并通过适当的耦合条件加以联系。考虑了三种耦合情况：LES-region在ans -region的上游，RANS应用在LES的上游，界面与流相切。法国和德国合作伙伴将共同开发所有三种类型的耦合程序，并首先对简单的通道流动进行测试。然后将这些方法应用于测试和进一步改进，以从该小组的常见测试案例中获得越来越复杂的流动，例如由德国合作伙伴进行的山丘，燃烧室和车体的流动，以及分离边界层，最后由法国合作伙伴进行的整个翼型的流动。