【摘要】 提出一种基于RBF神经网络多步预测的自适应PID控制算法.该算法用无局部极小的径向基函数网络对非线性系统进行在线辨识,利用多步预测误差对PID型控制器网络进行训练,从而实现PID参数的在线自适应寻优.通过对典型非线性系统的仿真研究,该控制系统具有较强的适应性和鲁棒性.更多还原

【Abstract】 An adaptive PID controller based on RBF neural network multi\|step prediction ispresented. The control algorithm identifies the parameters of nonlinear systems with radial trains the control neural network with a performance index basis function neural network,based on multi\|step prediction errors, and gets the optimal PID parameters on line. The results of simulation show that the proposed method is very effective to nonlinear system,and it achieves a better disturbance rejection performance and stronger robustness.更多还原