基于模拟退火策略的混沌优化算法的模糊神经网络控制器的设计  

Design of Fuzzy Neural Networks Controller Based on Chaos Optimization Algorithm of Annealing Strategy

作　　者：程启明[1] 

机构地区：[1]上海电力学院信控系,上海200090

出　　处：《测控技术》2004年第8期31-32,35,共3页Measurement & Control Technology

基　　金：上海市教委发展基金资助项目 ( 0 2 0K0 2 )

摘　　要：提出了一种基于模拟退火策略的混沌优化算法的模糊神经网络控制器 ,该模糊控制器由 4层BP神经网络表示 ,控制器参数由混沌粗搜索与细搜索相结合的优化策略优化。对非线性对象的仿真表明了优化方法的有效性和可行性 ,且控制具有无振荡、超调小、响应快、调节时间短、结构简单。A fuzzy neural networks controller based on chaos optimization algorithm of annealing strategy is proposed,it is expressed by four layers BP neural networks.The controller parameters are optimized by means of the combining strategy of global and partial chaos searching.The simulation that is applied to control a nonlinear system shows the validity and feasibility of optimal methods,and the system has some advantages of no oscillation,small overshoot,fast response,short setting-time,simple structure and easy realization.

关 键 词：混沌优化 模糊神经网络 最优控制 

分 类 号：TP2735[自动化与计算机技术—检测技术与自动化装置]