参数自整定PID控制器设计与仿真

对于工业控制领域中的系统普遍存在非线性、时变的特点，采用传统PID作为控制器很难获得满意的控制效果，而神经网络具有任意非线性逼近能力，可以通过对系统性能的学习来实现具有最佳组合的PID控制。设计了基于BP网络整定PID参数的控制器，该控制算法只需粗略给出PID参数便可以根据系统性能自动寻优调整。利用MATLAB软件得到的仿真结果表明，该控制策略可以达到满意的控制效果，且具有很强的鲁棒性和自适应能力。

Design and Emulation of Parameter Self-tuning PID Controller

In the field of industry control,the control result of traditional PID controller is not satisfied for the system which is non- linear an time-varying,since the ability of non-linear approximation,the neural network can accomplish the optimal combina- tion of PID controller by studying the system performance.This paper designs the parameter self-tuning PID controller based on BP network,the method needs only initial parameters of PID controller,and it can automatically tune according to the system performance.