近年来，我国在航空航天领域内取得重要进展，如四代机J20&31和大运Y20已首飞，大客C919明年首飞，超燃冲压发动机也备受关注。常规气动性能分析已成熟且为设计所用，但正面临着定常方法无法解决的非定常难题，如内埋弹仓、后舱门、起落架舱等开启，凹腔燃烧室稳定燃烧等。此时飞行器大部分区域附体，但局部非定常严重，迫切需要发展既能高效模拟附体流动且精细预测局部部件压力脉动、近声场和动载荷特性的高雷诺数湍流预测方法。 兼顾计算精度和效率的DES类方法预测大范围分离效果不错，已用于上述难题的简化模型分析；但因未含上游几何和流动特征，与真实存在差异。拟发展方法在附体区域用稀网格和RANS，分离区域用密网格和IDDES，通过植入的重叠区域获取上游信息并在其中用循环法自动获得解析湍流，克服其"灰区"难题。精选多个具有实验数据的标准算例进行方法的检验、验证和应用，高效地提供精细的非定常数据,指导飞行器设计。

Recently, many important progresses have been achieved in the field of aeronautics and astronautics in China. The new generation fighters, J-20 and J-31, and the large carrier Y-20 have finished their first flights; the large civil transporter, C-919, is planed to start to fly in the next year; the scramjet engine also draws great attention. The traditional analysis of the steady aerodynamics of aero-vehicles has matured and been used in airplane design. However, the designers are facing a difficulty that they can not well resolve the unsteady flow using traditional methods. The typical components are the inner weapon bay for the new generation military aircraft with stealth technology, aft door of the carrier, landing gear bay of the civil transporter, cavity of combustion chamber of the hypersonic vehicle and so on. The flows past those local components are locally massive separation, while most of the flow around the the aero-vehicles are attached. It's very urgent to develop highly efficient and highly accurate turbulence prediction methods for the high Reynolds number flows, using RANS to model the attched region and unsteady methods to resolve the pressure fluctuations, near-field noises and dynamic loads. DES-like methods, which combine high efficiency and accuracy, can well predict the massively separated flows and they have been successfully used to predict the flow past some simplified models. However, due to the absence of the features of upstream geometry and local flow, the flows past the simplified model are a little different from the real one. In this project, we are planning to develop an effcient and accurate turbulence prediction method for high Reynolds number flow, which applies the RANS methods to model the large scope of attached flow based on the coarse grids and uses the newest IDDES model to resolve the locally separated flow based on the fine grids. Between the RANS and IDDES regions, an overlap region are embedded, where the upstream features of geometry and flow can be introduced and the resolved unsteady turbulence can be generated automatically using Recycling algorithm, without any synthetic turbulence. It is named as Embedded-IDDES using Recycling algorithm in the Overlap Region (EIDDES-ROR). This method can address the problem of "grey region" of the DES-like methods, including the newest one, IDDES. Some test cases with plenty of experimental data are thoughtfully chosen to validate and verify the EIDDES-ROR method. The accurate and unsteady flow data are hoped to be effciently provided and it can be used to direct the aerodynamics and structure design.