电气工程学报 ›› 2017, Vol. 12 ›› Issue (4): 18-21.doi: 10.11985/2017.04.003

• 理论研究 • 上一篇    下一篇

基于RBF神经网络的电弧炉电极调节系统PID参数整定

鲁军,霍金彪   

  1. 沈阳理工大学自动化与电气工程学院 沈阳 110159
  • 收稿日期:2016-08-11 出版日期:2017-04-25 发布日期:2019-12-23
  • 作者简介:鲁 军 男 1965年生,博士,教授,研究方向为智能材料与智能控制系统。|霍金彪 男 1990年生,硕士,研究方向为控制理论与控制工程。

PID Parameter Setting on Electrode Adjustment System of Electric Arc Furnace Based on RBF Neural Network

Lu Jun,Huo Jinbiao   

  1. Shenyang Ligong University Shenyang 110159 China
  • Received:2016-08-11 Online:2017-04-25 Published:2019-12-23

摘要:

由于常规PID控制难以满足电弧炉电极调节系统复杂的工况,本文将径向基函数(RBF)神经网络与PID控制相结合,提出RBF-PID参数整定方法。通过RBF神经网络对控制对象Jacobian信息的辨识,采用增量式PID梯度下降算法整定已有的PID参数,设计了RBF-PID电极调节系统控制器。仿真结果验证了RBF-PID控制器能够实时整定PID参数,实现电极调节系统快速、准确地控制。

关键词: 电弧炉, 电极调节系统, RBF神经网络, PID控制, 参数整定

Abstract:

Because conventional PID control is difficult to meet the complex conditions for electrode adjustment system of electric arc furnace, the radial basis function (RBF) neural network is introduced into the PID control, a setting method of RBF-PID is proposed in this paper. By RBF neural network identifies Jacobian information of control object, selecting incremental PID gradient descent algorithm sets existing PID parameters, a RBF-PID controller of electrode regulating system is designed. The simulation results verify that RBF-PID controller can real-time set the PID parameters, realize quickly and accurately control to electrode regulating system.

Key words: Electric arc furnace, electrode adjustment system, radial basis function neural network, PID control, parameter setting

中图分类号: