| 一种基于RBF神经网络预测模型的DMC预测控制 |
| |
| 引用本文： | 李志武. 一种基于RBF神经网络预测模型的DMC预测控制[J]. 科学技术与工程, 2010, 10(7) |
| |
| 作者姓名： | 李志武 |
| |
| 作者单位： | 辽宁石油化工大学职业技术学院,抚顺,113001 |
| |
| 摘 要： | 研究了一种利用RBF神经网络预测模型的动态矩阵控制算法,首先利用动态节点生成构造性RBF神经网络辨识对象模型,同时预测对象的未来输出,然后用传统的动态矩阵控制算法进行滚动优化和反馈校正。仿真表明该算法在非线性对象的任意工作点都可以通过神经网络辨识获得工作点附近的近似线性模型,具有较好的实时性。
|
| 关 键 词： | 径向基神经网络 动态矩阵 动态结点生成训练算法 非线性模型预测 |
| 收稿时间： | 2009-11-30 |
| 修稿时间： | 2009-12-22 |
Dynamic Matrix Predictive Control Based on RBF Neural Networks Predictive Model |
| |
| li Zhiwu. Dynamic Matrix Predictive Control Based on RBF Neural Networks Predictive Model[J]. Science Technology and Engineering, 2010, 10(7) |
| |
| Authors: | li Zhiwu |
| |
| Affiliation: | Vocational and Technological College/a>;Liaoning Shihua University/a>;Fushun 113001/a>;P.R.China |
| |
| Abstract: | A new Dynamic Matrix Control(DMC) algorithm using Radial Basis Function(RBF) neural networks as predictive model is proposed.First the dynamic node creating training RBF neural network is adopted for identifying model of the controlled plant and produce predictive output.Then the DMC algorithm is used in receding horizon and revising feedback.After the proposed algorithm is applied to simulate a class of nonlinear plants,the result demonstrates that the approximate linear model near work spot can be achieve... |
| |
| Keywords: | RBF neural networks DMC dynamic node creation training algorithm nonlinear model prediction |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《科学技术与工程》浏览原始摘要信息 |
|
点击此处可从《科学技术与工程》下载全文 |
