连续可微函数全局优化的混合遗传算法

A Hybrid Genetic Algorithm for Global Optimization of Continuous-differential Function

通过在遗传算法(GA)中定义最速下降(SD)算子、适应度和数据结构,从而得到结合GA和SD法长处,既有较快收敛性,又能以较大概率求得连续可微函数全局极值的混合遗传算法。数值结果表明该方法优于GA和SD法。

Through defining a steepest decent operator, a fitness, and numerical structure in the genetic algo-rithm, a hybrid genetic algorithm for global optimization of continuous -differential function, combined the advances of both genetic algorithm and steepest decent algorithm, is got with the faster convergence and the greater probability. The numerical results show that the method is suprerior to the genetic algorithm and the steepest decent algorithm.