首页 | 本学科首页   官方微博 | 高级检索  
     

基于改进BP神经网络的PID控制方法研究
引用本文:史春朝,张国山.基于改进BP神经网络的PID控制方法研究[J].计算机仿真,2006,23(12):156-159.
作者姓名:史春朝  张国山
作者单位:天津大学电气与自动化工程学院,天津,300072
摘    要:针对最速下降法收敛速度慢和易陷入局部极小的缺点,提出一种新型的基于改进BP神经网络的PID控制方法,该方法将神经网络和PID控制策略相结合,既具有神经网络自学习、自适应及逼近任意函数的能力。又具有常规PID控制器结构简单的特点。该控制器的算法采用Fletcher—Reeves共轭梯度法,它可以避免网络陷入局部极小点,同时加快网络的训练速度。并用这种改进的共轭梯度法对神经网络PID控制器参数实现在线修正。最后给出了在Matlab平台上的实现算法。仿真结果表明该控制方法是有效的。

关 键 词:误差反传神经网络  改进共轭梯度法  比例-积分-微分控制
文章编号:1006-9348(2006)12-0156-04
收稿时间:2005-10-20
修稿时间:2005年10月20

Study of PID Control Based on Improved BP Neural Network
SHI Chun-chao,ZHANG Guo-shan.Study of PID Control Based on Improved BP Neural Network[J].Computer Simulation,2006,23(12):156-159.
Authors:SHI Chun-chao  ZHANG Guo-shan
Affiliation:School of Electrical and Automation Engineering, Tianjin University, Tianjin 300072, China
Abstract:To deal with the defects of the steepest descent in slowly converging and easily immerging in partial minimum, this paper proposes a new type of PID control method based on the improved BP neural network, which is a combination of the neural network and the PID strategy. It has the merits of both neural network and PID controller. Moreover, Fletcher - Reeves conjugate gradient in controller can make the training of network faster and can eliminate the disadvantages of steepest descent in BP algorithnL The parameters of the neural network PID controller are modified on line by the improved conjugate gradient. The programming steps under MATLAB are finally described. Simulation result shows that the controller is effective.
Keywords:Back-propagation(BP) neural network  Improved conjugate gradient  Proportional-integral-derivative(PID) control
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号