柔性多面体搜索算法在遗传算法中的综合应用研究

首先分析了柔性多面体搜索算法和基本遗传算法两者结合的基础，提出了柔性多面体方向进化算子和柔性多面体交叉算子，以这两个新的遗传操作算子和柔性多面体搜索算法为基础，提出了两种新的混合遗传算法FP-HGA(Ⅰ)、FP-HGA(Ⅱ).在FP-HGA(Ⅰ)中，嵌入了柔性多面体方向进化算子和柔性多面体搜索算法；在FP-HGA(Ⅱ)中，嵌入了柔性多面体交叉算子，用FP-HGA(Ⅰ)、FP-HGA(Ⅱ)及SGA(Simple Genetic Algorithm)来求解Rosenbrock测试函数的最小值，FP-HGA(Ⅰ)和FPHGA(Ⅱ)算法和SGA算法的计算结果表明该混合遗传算法在收敛速度和精度方面均得到很大提高．

Research on the Synthetic Applications of Flexible Polyhedron Search Algorithm to Simple Genetic Algorithm

In this paper, first analyze the foundations of fusing flexible polyhedron search algorithm into simple genetic algorithm(SGA)and propose two new genetic operators, one is the flexible polyhedron directional evolution operator, another is the flexible polyhedron crossover operator. Secondly, by use the two new genetic operators, construct two new hybrid genetic algorithms called FP _HGA(I)and FP _HGA(II), the flexible polyhedron directional evolution operator and the flexible polyhedron crossover operator are embedded into the FP _HGA(I)and FP _HGA(II)respectively. Third, verify the three genetic algorithms of SGA, FP _HGA(I)and FPSA _HGA(II)by using them to obtain the minimum of the experiment function of Rosenbrock, the results indicate that FP _HGA(I)and FP _HGA(II)have better performance indexes than SGA does in improving convergence speed and calculation precise and FP _HGA(I)has the best performance indexes.