基因算法在喷管反设计中的应用

在气动优化设计中引进了基于达尔文的自然选择机制和生物遗传机制的基因算法，构造了相应的基因变异等算子。针对喷管气动外形的特点提出了与之相适应的Ｂéｚｉｅｒ曲线基因表达，把算法搜索空间缩小为由有限参数曲线控制点构成的子空间，从而加快了算法收敛速度。并与全位势方程求解程序相结合，提出了具体的供优化设计的适应函数，研制了二维气动反设计计算程序，进行了二维喷管反设计数值实验。结果表明，本文构造的方法是可行的，而且具有全局优化和鲁棒性强等特点

Using Genetic Algorithms for Optimum Shape Design of Nozzle

Applys Genetic A lgorithms (GAs), based on the mechanics of natural selection and gene tics to the optimization of aerodynamic design. The genetic operators are constr ucted. Bézier curve representation is proposed based on the features of the ae rodynamic shapes and the principle of genetic algorithms. Bézier curve represen tation can reduce the searching scope of GAs because it needs only small control po ints to describe an aerodynamic shape. The fitness function is defined based on the solution of full potential equation. A two-dimensional program for the opti mization of aerodynamic design is developed. Then the method is applied to nozzl e reconstruction. The numerical experiments are carried out for the design of 2 -D aerodynamic nozzle. The results show that the method is feasible and has the abil ity to obtain the approximate optimum shape within finite genetic generations.