Internal Model Control of Three-Phase Four-Leg Active Power Filter Based on Online LS-SVM ANN Inverse System Method

Abstract:

In dealing with the problem of the SAPF’s nonlinear and strong coupling model, the internal model control of three-phase four-leg active power filter based on online ANN method is studied in this paper. With the ANN’s nonlinear mapping ability of self-learning and self-organizing modeling, the inverse system can be approximated by online LS-SVM. In order to have a good linearization control effect, the internal model control based on ANN is proposed for the combined pseudo-linear system. This method can be used to design effective controllers for nonlinear system with unknown mathematical models. At last, the simulation results show that a good steady-state performance can be obtained under the improved method

• Periodical: Advanced Materials Research (Volumes 383-390)
• Pages: 2132-2137
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.383-390.2132
• Online since: November 2011
• Authors: Hong Qi, Zhen Hua Shao
• Keywords: Artificial Neural Network (ANN), Inter Model Control (IMC), Inverse System Method, Least Squares Support Vector Machines (Ls-SVM), Shunt Active Power Filter (SAPF)

References

[1] F.Z. Peng and D.J. Admas, Harmoonics sources and filtering approaches, , in Proc. Ind. Appl. Conf., vol. 1, Oct. 1999, pp.448-455.

[2] Patricio Salmeron , and Salvador P. litran, A control strategy for hybird power filter to cpmpensate four-wires three-phase systems, IEEE Transaction on Power Elecectronics, vol. 25, July 2010, p.1923-(1931).

DOI: 10.1109/tpel.2010.2043687

[3] Xiuchong Liu, Huaguang Zhang, Hongzhi Chen. Control strategy of fourth leg in four-leg inverter, (in chinese)Proceedings of the CSEE , vol. 27 , 2007, pp.87-92.

[4] Li Qiao, Active Disturbance Rejection Control and H∞ Control for Shunt Active Power Filter. (in chinese) Guangdong, South China University of Technology, (2005).

[5] Ge BaoMing, Wang XiangHeng, Su PengSheng, et al. Nonlinear internal-model control for switched reuctance drives,. IEEE Transaction on Power Electronics, vol. 17, March 2002, pp.379-388.

DOI: 10.1109/tpel.2002.1004245

[6] Suykens J, Vandewale J. Least square support vector machine classifiers, Neural Processing Letters, vol. 9, March 1999, pp.293-300.

[7] Liu Bo, Hao Zhifeng, Yang xiaowei. Nesting support vector machine for multi-classificating, Proceedings of the Four international Conference on Machine Learing and Cybernetics, Guangzhou, 2005, pp.4220-4224.

DOI: 10.1109/icmlc.2005.1527678

[8] Lei Bicheng, Wang Wanliang, Li Zuxin. Networked predictive control system based on online LS-SVM, (in chinese) Information and control, vol. 38, April 2009, pp.163-169.

DOI: 10.1109/wcica.2008.4594157

[9] Mei Shen-wei, Shen Tie-long, Liu Kang-zhi. Modern robust control theory and application. Beijing: Tsinghua University Press , (2007).

[10] Zhenhua Shao, Chong Chen, Ruiquan Lin. A three-phase active power filter based on improved adaptive harmonic detection,. Asia-Pacific Power and Energy Engineering Conference 2010. Chengdu, China. March 28-31, (2010).

DOI: 10.1109/appeec.2010.5449292