用进化RBF神经网络控制Lorenz混沌

对镇定一嵌入在Lorenz混沌吸引子内的不稳定平衡点上的混沌轨道提出了一种利用进化RBF网控制混沌系统的新方法，采用了基于两层编码改进进化规划（IEP）的RBF网学习算法，要以同时确定网络的拓扑结构和参数，仿真结果表明本文控制Lorenz混沌响应速度快，控制精度高。

CONTROL OF LORENZ CHAOS USING EVOLUTIONARY RBF NEURAL NETWORK

A new method of chaotic control with evolutionary RBF neural network is presented for stabilizing chaotic orbits on an unstable equilibrium embeded within a Lorenz chaotic attractor. A learning algorithm of RBF neural network based on improved evolutionary programming (IEP) with two-layer encoding has been used, the algorithm simultaneously determines both the topology and parameters for RBF neural network. The simulation results have shown that rapid response and higher accuracy of the Lorenz chaotic control can be obtained.