论文部分内容阅读
本文提出一类基于高斯基神经网络的自学习控制器,该控制器由两个GPFN网络组成,一个完成PID学习控制,另一个完成未知被控对象模型的建模.为加快网络的学习过程,文中提出了递归最小二乘法(RLS)用于神经网络的学习,并分析研究了自学习控制系统的收敛性和稳定性.仿真和实验结果表明,这类智能控制是成功的.
In this paper, a kind of self-learning controller based on Gaussian neural network is proposed. The controller consists of two GPFN networks. One is to complete the PID learning control and the other to complete the modeling of the unknown controlled object model. In order to speed up the process of network learning, a recursive least squares (RLS) algorithm is proposed for neural network learning. The convergence and stability of the self-learning control system are analyzed. Simulation and experimental results show that this kind of intelligent control is successful.