基于神经网络多模型自适应切换控制研究

针对传统自适应控制和现有多模型自适应切换控制理论和方法中存在的问题，提出面向复杂系统的神经网络多模型自适应切换控制方法。采用最近邻聚类学习算法对样本分类，并利用RBF神经网络的学习能力和非线性逼近能力进行离线建模。同时，引入动态模型库技术来动态建立多模型。系统运行时在线实时检测系统当前状态，若超出现有各子模型所构成的状态空间，则利用在线神经网络学习新状态并建立新模型加入动态模型库中，从而改善动态系统的暂态响应和增强系统的鲁棒性。计算机仿真结果，表明了该算法的有效性。

MULTI-MODEL ADAPTIVE SWITCHING CONTROL BASED ON NEURAL NETWORK

A multi-model adaptive switching control based on neural network is proposed,which aims at the problem of conventional adaptive control and multi-model adaptive control in existence.We adopt the closest clustering algorithm to classify the samples,and then utilize the learning and nonlinear approximate abilities of the RBF neural network to construct an off-line model.Meanwhile dynamic model bank technology is applied to establish multi-model.If the status of real time system go beyond the space,which is formed of all existing sub-model,a new status is learnt through on-line neural network and a new model,which is added in the dynamic model bank in order to improve the dynamic system's transient response and robustness,is set up.Finally,simulations are given to demonstrate the validity of the proposed method.