摘要
采用类别形状函数变换(CST)方法对翼型参数化描述;利用Fluent软件,进行翼型升力系数和阻力系数的气动计算;以升阻比最大为优化目标,通过拉丁超立方设计生成样本点,建立了径向基神经网络(RBF)代理模型;使用粒子群优化算法(PSO),在Isight平台上,实现对Clark Y翼型优化的整个过程。优化翼型升阻比比原始翼型提高了约10%,表明此种方法是可行的,可用于小型无人机设计的工程中。
The CST universal parametric geometry representation method was used to describe the parametric air- foil, and lift coefficient and drag coefficient of the designed airfoil was directly calculated by Fluent. The maxi- mum lift-drag ratio as the goal was achieved by employing PSO algorithm. The latin hypercube method was em- ployed to construct the initial sample points used in a RBF surrogate model. The software of Isight was used to realize the design processes. The maximum lift-drag ratio was significantly improved by about 10%. The result showed that this approach was feasible and could be used as reference for Mini-UAV design in engineering.
出处
《海军航空工程学院学报》
2013年第3期307-310,322,共5页
Journal of Naval Aeronautical and Astronautical University
