针对纯滞后系统的RBF神经网络智能控制的研究

针对工业控制过程中普遍存在的大惯性、纯滞后、时变性、非线性对象的控制问题,采用传统的控制方法不能达到满意的控制效果,提出了基于RBF神经网络的PID自适应控制方案。采用神经网络辨识器在线辨识系统模型,自动调整PID控制器参数,从而实现系统的智能控制。仿真结果表明：该方法对于纯滞后控制系统能进行有效的控制并且具有很好的自适应性和鲁棒性。

Research on the RBF neural network-based intelligent control for improving a control system with time delay

The conventional control method can＇t acquire a satisfying result from any industrial process control system with big inertia, pure time edlay, nonlinearity and time variation. The paper presents a self-adaptive PID control strategy based on a neural network provided with an identifier to follow up the system model in an on-line way and adjust the PID control parameters, thus achieving the intelligent control of the system. The simulation result shows that this control method is effective to the control systems with time delay and is of great adaptability and robustness.