基于模糊神经模型的自适应单神经元控制系统设计

Adaptive Single-Neuron Control System Design Based on Fuzzy Neural Network Model

提出一种基于模糊神经模型的自适应单神经元控制系统。该控制系统首先根据采集到的输入输出数据建立被控过程模型,并在此基础上引入单神经元控制器。通过李亚普诺夫方法对控制器参数进行在线调节,从而使得系统输出值能够较快跟踪设定值。理论分析和仿真结果表明:本文提出的单神经元控制器和传统的PID控制器具有极其相似的结构,因此,具有结构简单、易于操作的特点,具有较快的跟踪速度,并且控制参数可以在线调节。

An adaptive single-neuron control system based on neuron-fuzzy model is proposed in this paper. Firstly, the nonlinear process model is identified by input-output points. Then the single-neuron controller, which is adjusted using the Lyapunov method, is considered so that the setpoint can be rapidly tracked by the output of the system. Theory analysis and simulation results show that the proposed singleneuron controller mimics the conventional PID controller. Consequently, it possesses simple structure and can be easily operated . Moveover, this adaptive single-neuron controller is better than conventional PID controller, and the parameters of the controller are on-line adjusted.