Optimization of Fractional Order Controller Parameters

Article Preview

Abstract:

A new training method is proposed, which could solve the problem of that parameters of fractional order controller are not easy to be selected. This method which based on the principle of gravity optimizes parameters. Random initial parameter based on step was set as coordinate form which in the midpoint of the multidimensional space. The error between the actual output and the target output was set as radius. This method had advantages which could not need to calculate the gradient and could randomly select initial. Through the simulation experiment, this method is successfully applied in the fractional order PID controller, which obtains the optimal parameters.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

1620-1623

Citation:

Online since:

December 2014

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2015 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] GAO Xue-peng, CONG Shuang. Comparative Study on Fast Learning Algorithms of BP Networks[J]. CONTROL AND DECISION, 2001, 16(2): 167-171.

Google Scholar

[2] JIANG Peng-fei, CAI Zhi-hua. Combined algorithms for training RBF neural networks based on genetic algorithms and gradient descent[J]. Computer Applications4, 2007, 27(2): 366-369.

Google Scholar

[3] XUE Ding-yu, ZHAO Chun-na. Fractional order PID controller design for fractional order system[J]. Control theory and Applications, 2007, 24(5): 771-776.

Google Scholar

[4] ZHANG Yi, MEI Feng-xiang. Fractional Differential Equations of Motion in Terms of Riesz Fractional Derivatives[J]. Transactions of Beijing Institute of Technology, 2012, 32(7): 766-769.

Google Scholar