一种模糊神经网络控制器参数的混沌优化设计

通过模糊控制与神经网络相串联的方式构成模糊神经网络系统，然后提出一种基于模拟退火策略的混沌优化算法，将该算法引入模糊神经网络参数域中进行优化，实现混沌粗搜索与细搜索相结合优化目的，体现出具有更强的模糊神经网络参数全局最优解的搜索能力。采用该控制器对一个非线性对象进行控制。仿真实验表明，该方法能有效地实现模糊神经网络控制器参数优化，控制具有无振荡、超调小、调节时间短等优点，算法结构简单，容易实现。

Optimization design of fuzzy neural networks controller parameter based on chaos

A fuzzy neural networks controller, which is constructed by cascade of fuzzy logical and feedforward neural networks, is introduced. Based on chaotic variable and annealing strategy, an optimization method is proposed to search for global optimal fuzzy neural networks controller. Controller parameters are optimized by means of combination of global and partial chaos searching.Because chaos has no repetition and randomicity, it has higher ability of searching for optimal fuzzy neural networks. The method is applied to control a nonliner system. Simulation results show that the optimization method is efficient to search for controller parameters, and the control has advantages of small overshoot,short setting time.The algorithm has simple structure and is easy to realize.