基于KOHONEN神经网络的模拟退火算法  

A Simulated Annealing Algorithm based on KOHONEN Neural Network

作　　者：谭建辉[1] 

机构地区：[1]广东省阳江职业技术学院,阳江市529500

出　　处：《微计算机信息》2006年第02Z期254-256,共3页Control & Automation

基　　金：国家"十五"国防预研基金的资助

摘　　要：本文提出了一种基于分布式知识的广义优化方法----基于KOHONEN神经网络的模拟退火算法。采用人工神经网络积累搜索区域信息,利用其学习的分布式寻优知识指导搜索过程的优化。研究了各种优化策略,给出了算法流程和结构框图。典型算例的仿真研究结果显示了新算法在寻优性能上的优越性。A generalized optimization method based on distributed knowledge simulated annealing algorithm based on KOHONEN neural network is presented in this paper. The key point of this method is to accumulate the information of the solution space using KOHONEN neural network and to guide the further search according to the distributed knowledge studied. In this paper, various optimization maneuvers are discussed, and the flow chart of the method is given. Computational results on typical optimization problems are used to demonstrate the superiority of this method in optimization performances.

关 键 词：神经网络 模拟退火 广义优化 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]