基于模矢搜索和遗传算法的混合约束优化算法

摘要

近年,免梯度方法又开始引起大家的注意,由于不需要计算函数的梯度,特别适合用来求解那些无法得到梯度信息或需要花很大计算量才能得到梯度信息的问题.本文构造了一个基于模矢搜索和遗传算法的混合优化算法.在模矢搜索方法的搜索步,用一个类似于遗传算法的方法产生一个有限点集.算法是全局收敛的.%Recently,so-called derivative-free methods have attracted much attention,which don't require computation of derivatives of function and are particularly suitable for problems which the derivatives are not available or are extremely expensive to compute. This paper presents a hybrid algorithm which combines the pattern search method and the genetic algorithm for bound constrained optimization. In the search step of the patter search algorithm,a finite set of points is obtained by a process similar to genetic algorithm. In theory,the algorithm is globally convergent.