跨音速压气机轮毂型线优化设计

摘要

Setting Rotor37 as the research object,the optimization design of hub- shape contouring is carried out by using the Design3D module of Numeca software. This module combining CFD、artificial neural network and genetic algorithm could be applied to 3D aerodynamic optimization design. Comparing with the original hub - shape contouring, the optimized Rotor37 has better performance with higher isentropic efficiency and the total pressure ratio. By analysing the changes of geometry, overall performance and flow field in detail, it shows that the improvement mainly results from the forth movement of shock and the control of boundary layer separation.%针对跨音速转子Rotor37,采用Numeca软件中的Design3D模块对其叶片区域的轮毂型线进行了优化设计计算.模块结合了CFD、人工神经网络和遗传算法,优化效果显著.轮毂型线优化后的Rotor37相比原始设计,在整个设计转速范围内的等熵效率得到了明显提高,而裕度与优化前相比略微提高.通过流场分析,其性能的改善主要源于叶片根部区域的激波前移和尾缘处附面层分离的减弱.