多元多峰值目标函数在MATLAB中的优化  

Multivariate multimodal function optimization based on MATLAB

作　　者：姚运萍[1,2] 王倩倩[1,2] 

机构地区：[1]兰州理工大学数字制造技术与应用省部共建教育部重点实验室,甘肃兰州730050 [2]兰州理工大学机电工程学院,甘肃兰州730050

出　　处：《微型机与应用》2012年第10期78-80,共3页Microcomputer & Its Applications

摘　　要：采用MATLAB的遗传算法,利用强大的数学计算能力和遗传工具箱,在全局搜索空间内寻找极值点,能够有效地对多元多峰值函数进行优化,避免了利用传统优化方法在多元多峰值函数优化过程中陷入局部极值点的优化误区。同时简要介绍了遗传算法的特点、流程和优化工具箱,通过实际编程优化,说明基于MATLAB的遗传算法是一种具有良好的全局寻优的优化工具。Muhivariate muhimodal functions have multiple local extreme points. However, it is easy to fall into a single local extreme point through traditional optimization methods and it＇s difficult to meet the global optimization requirements. Genetic algo- rithm, which has powerful mathematical computing capability, can search the global space to find extreme points and can effectively optimize multivariate muhimodal functions. And a wrong conclusion when traditional optimization methods were used in optimization of multivariate muhimodal functions can be avoided. Processes and optimization toolbox of genetic algorithm are briefly introduced in this paper. It indicates that the MATLAB-based genetic algorithm is a good global optimization tool by actual programming optimization.

关 键 词：遗传算法 全局寻优 多元多峰值函数优化 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]