并行进化算法及其在航空工程优化问题中的应用

提出了用于加速气动力形状优化过程的分级型Nash基因算法．分级型算法可以看作是并行基因算法的特例，后者使用了互相联系但独立进化的子群的概念．本文在并行基因算法中引入多层分级拓扑结构以提高算法的收敛性．这种拓扑结构混合使用不同精度的模型，低精度模型用于探索搜索空间，高精度模型用于对准优解进行提纯．将此方法与Nash博弈相结合，构造了多目标优化算法，并应用于气动力优化问题．针对喷管反设计问题与多段翼型高升力优化问题，在计算机集群并行环境下进行了计算，结果表明本文的算法具有较高的加速收敛特性．

Parallel Evolutionary Algorithms for Optimization Problems in Aerospace Engineering

New Hierarchical Genetic Algorithms (HGAs) mixing Nash game are presented to speed up the optimization of aerodynamic shapes. The approach of HGAs is a particular instance of Parallel GAs (PGAs) based on the notion of interconnected sub-populations evolving independently. To improve convergence history, the present study introduces a multi-layered hierarchical topology in PGAs. Such a topology allows the use of multiple models for optimization problems, and shows that it is possible to mix fast Low Fidelity models for exploration and expensive High Fidelity models for exploitation. Finally, a new class of multi-objective optimizers mixing HGAs and Nash game is defined. These methods are tested for solving design optimization problems in Aerodynamics. A parallel version of this approach running on a cluster of PCs demonstrates the convergence speed up on an in verse nozzle problem and a high lift problem for a multiple element airfoil.