The Application of RBF-NN with Improvements in Clustering Algorithm Based on Ant Colony Optimization in PID Control

Author

Zhang, Hua ; Kong, Feng ; Fu, Xiuwei ; Zhang, Dongdong

Author_Institution

Dept. of Electron. Inf. & Control Eng., Guangxi Univ. of Technol., Liuzhou, China

Volume

2

fYear

2009

fDate

12-14 Dec. 2009

Firstpage

266

Lastpage

269

Abstract

The PID control of RBF-NN is taken for the nonlinear system. Due to the low quality of clustering in the clustering algorithm of the traditional RBF-NN, the rate of convergence is directly influenced by the initial value. In this paper, the quality of the clustering has been raised and the initial value has been optimized through the improvements of the clustering algorithm by taking the K-means-algorithm and the Ant Colony Optimization (ACO). The simulation results show that the rate of convergence is precise and fast after the clustering algorithm is improved and the PID control is better than the one without taking the new method.

Keywords

convergence; nonlinear systems; optimisation; pattern clustering; radial basis function networks; three-term control; K-means algorithm; PID control; RBF-NN; ant colony optimization; clustering algorithm; convergence rate; neural network; nonlinear system; radial basis function; Algorithm design and analysis; Ant colony optimization; Application software; Clustering algorithms; Computational intelligence; Control systems; Convergence; Intelligent control; Nonlinear control systems; Three-term control;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Design, 2009. ISCID '09. Second International Symposium on

Conference_Location

Changsha

Print_ISBN

978-0-7695-3865-5

Type

conf

DOI

10.1109/ISCID.2009.213

Filename

5368436