基于BP算法PID控制器的研究

文中主要研究了基于BP算法的PID控制器在非线性系统中的控制效果以及对权值整定初始化的优化。在介绍BP网络基本原理的基础上以非线性控制系统Simulink仿真为例,使用基于BP算法的PID控制器对该系统进行优化和整定,并结合Nguyen-Widrow初始化算法为层产生初始权重和偏置值,使得每层神经元的活动区域能大致平坦地分布在输入空间,实现了对PID参数的实时调节,并且使神经网络的学习和收敛速度加快,大大改善系统的初始运行的稳定性。仿真结果表明,基于BP算法的PID控制器在非线性控制系统中对其参数优化整定具有良好的效果。

Research of PID Controller Based on BP Algorithm

Studies the effects of the BP algorithm based on PID controller in the nonlinear system as well as setting the right values to initialize the optimization.In this paper take example for Simulink simulation of a non-linear control systems,use PID controller based on BP algorithm to optimize the system and setting,which combined with Nguyen-Widrow initialization algorithm,realize the real-time adjustment of the PID parameters,and accelerate the neural network learning convergence.Simulation results show that ...