叶轮机械是航空发动机和燃气轮机的核心部件。叶轮机设计的一个关键问题就是准确预测叶片边界层由层流向湍流的转捩过程。本课题旨在发展适用于叶轮机械流动转捩/湍流的高效精细模拟方法，进而通过该方法揭示具有高负荷三维叶型的叶轮机失速机理，为先进叶轮机的气动设计提供理论支撑。具体研究拟分为三个阶段：1）应用大涡模拟（LES）方法研究典型平面叶栅流动转捩/分离的非定常演化规律，继而利用动态模态分解和频谱分析方法处理高精度流场数据，获得多重转捩模态的特征频率与能量分布规律；2）根据转捩模态特征，改进申请人原有的叶轮机械流动转捩/湍流模式以及RANS/LES混合方法，并将二者耦合，发展适用于叶轮机械流动转捩/湍流的高效精细模拟方法；3）针对存在动静干涉的单级叶轮机,应用该方法开展全周流场模拟，掌握其失速特征与关键参数。

Turbomachines are the core components of both aeroengines and gas turbine engines. Laminar-turbulent transition prediction plays a key role in the design of turbomachinery blades. The present work is to develop numerical methods of both high fidelity and efficiency, to reveal flow transition/separation mechanisms on real-geometric blades, and further rotating stall mechanisms for highly-loaded axial turbomachines. We divided this work into three stages: 1) large eddy simulations (LES) will be conducted of a typical cascade flow transition/separation; subsequently, dynamic mode decomposition and spectrum analysis will be used to extract the characteristic frequencies of transition modes as well as their energy distributions; 2) based on the obtained transition-mode characteristics, a transition/turbulence model available in turbomachinery flows will be proposed and validated, then coupled with a hybrid RANS/LES method; 3) Consequently, with the aid of the new method, we will simulate a full annulus of a turbine stage with rotor-stator interaction, in order to obtain the key parameters in stall-inception processes.