A Study of Multi-Objective Aerodynamic Optimization Design for Variable Camber Airfoils and High Lift Devices

变弯翼型与增升装置多目标气动优化设计研究

Received: 28 April 2017

Abstract

Aiming at the synthetical optimization of the aerodynamic performance between the low-speed condition of two-dimensional high lift devices during take-off and landing phase and the high-speed condition of variable camber airfoil during cruise phase, an aerodynamic optimization design method for high lift device based on Kriging based surrogate model and multi-objective genetic algorithm has been developed. With the application of Adaptive Dropped Hinge Flap mechanism, the low-speed take-off and landing performance and high-speed cruise performance of the aircraft is improved by coupling deflection of the flap and spoiler. The position of flap hinge, deflection angle of spoiler and deflection angle of flap are taken as design variables; The Navier-Stokes equations are used to predict the aerodynamic forces of initial samples; The Kriging based surrogate model is employed to establish the algebraic relation between design variables and aerodynamic forces at take off, landing and cruise, obtaining four efficient prediction models for aerodynamic forces; Multi-objective optimization design with multi-objective genetic algorithm is conducted on the basis of surrogate models. The automatic generation of computational grid is achieved by the mesh deformation method based on RBF (Radial Basis Function) when the design variables change. On the basis of efficient global multi-objective optimization design platform, the synthetical optimization of high-speed and low-speed aerodynamic performance is conducted; The multi-objective solution set of the Pareto frontier is verified and analyzed, and the optimal solution with well matched high and low speed performance is selected.

摘要

针对二维增升装置低速起飞、着陆状态和变弯度翼型高速巡航状态下的高低速气动特性进行综合优化研究，发展了一种基于Kriging代理模型与多目标遗传算法的增升装置多目标气动优化设计方法。使用自适应下垂式铰链襟翼机构，通过襟翼与扰流板联动偏转来改善飞机的低速起降性能及高速巡航性能。将襟翼铰链位置、扰流板偏角和襟翼偏角作为设计变量，通过求解N-S方程来预测初始样本点气动力，利用Kriging代理模型分别建立起飞、着陆和巡航状态下设计变量与气动力之间的关系，得到多个高效气动力预测模型，最后使用多目标遗传算法在代理模型的基础上进行多目标气动优化设计。设计变量变化时计算网格的自动生成采用RBF（radial basis function）动网格技术来实现。基于搭建的高效全局多目标优化设计平台进行了高低速综合气动性能多目标优化，并对Pareto前沿面的多目标解集进行了多设计点校验与分析，筛选出了多目标并重的优化解。