采用遗传算法的并行最优统计设计

Parallel Optimal Statistical Design Using Genetic Algorithms

对于最优统计设计，本文提出了遗传算法（ＧＡ）以及边界抽样策略。所提出的方法，以改善的计算速度，为响应表面函数近似的选择提供了一个满意的模型代表。此外提出了并行ＧＡ操作的两种可能的方法。

Genetic algorithms (GA) together with a boundary sampling strategy are proposed for optimal statistical design. The proposed approach can provide a satisfactory presentative model at improved computation speed ofthe selection of the response surface function approximation.In addition,two possible approaches to parallelling the GA operations are proposed.