基于自适应Kriging代理模型的叶型气动优化设计

Aerodynamic Optimization Design of Airfoils Based on Adaptive Kriging Surrogate Model

基于自适应Kriging代理模型的近似代理技术,采用Bézier函数实现叶型几何参数化,并结合CFD性能求解,建立了叶型优化设计方法,改进了该代理方法的寻优过程,通过Bézier控制点实现叶型几何变化的控制,自动完成某涡轮叶型头部型线的优化设计.结果表明:优化后的叶型压力恢复系数有所增大,提高了叶型的性能,验证了该优化系统的可靠性,提高了叶型优化的效率.

An aerodynamic optimization design for turbine airfoil was developed using adaptive Kriging sur- rogate model, during which the airfoil geometry parameterization was realized with B4zier function and CFD method. Searching process of the surrogate model was improved, and the airfoil geometry was modi- fied through controlling via Bezier control points, so as to automatically achieve optimization design of the leading edge of turbine blade. Results show that that higher pressure recovery coefficient and better aero- dynamic performance are obtained after aerodynamic optimization of the airfoil, which proves the optimiza- tion model to be reliable and effective.