总体优化一类双参数填充函数算法的改进  

Modification of a Class of Double-Parameterized Filled Function Methods for Global Optimization

作　　者：朱文兴[1] 

机构地区：[1]福州大学计算机系

出　　处：《数学物理学报（A辑）》1999年第S1期550-558,共9页Acta Mathematica Scientia

基　　金：国家973项目!<数学机械化与自动推理平台>;福建省自然科学基金

摘　　要：求解无约束总体优化问题的一类双参数填充函数算法需要假设该问题的局部极小解的个数只有有限个，而且填充函数中参数的选取与局部极小解的谷域的半径有关．该文对其填充函数作了适当改进，使得新的填充函数算法不仅无需对问题的局部极小解的个数作假设，而且填充函数中参数的选取与局部极小解的谷域的半径无关．数值试验表明算法是有效的．A class of double-parameterized filled function methods developed for unconstrained global minimization problem needs the assumption that the optimization problem has only a finite number of local minimizers, and has parameters which are restricted by the minimal radius of the S-basin of some local minimizer of the problem. In this paper, we modify them such that the two weaknesses are overcomed. Numerical experiments show that the algorithm is efficient.

关 键 词：总体优化 双参数填充函数算法 局部极小解的个数 

分 类 号：O241[理学—计算数学]