Predictive Control of DC/DC Converter Based on State Observer

Article Preview

Abstract:

In this paper, we present an advanced way to control the DC/DC converter by using predictive control. As the current state of the circuit must be known while using the predictive control, state observer is applied to solve the problem that some variables of DC/DC converter can not be observed. Neural network optimization is used to solve the QP problems in single sample step of predictive control. Simulation results show that this approach can utilize fast converge property of neural network and the new control strategy turns out to be very efficiency.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 476-478)

Pages:

542-546

Citation:

Online since:

February 2012

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2012 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] P.Dananjayan,V.sriRam,C.Chellamuthu:A flyback constant frequency ZCS and ZVS quasi-resonant converter, Microelectronics Journal. (1998), pp.495-504.

DOI: 10.1016/s0026-2692(97)00110-9

Google Scholar

[2] I. Koudar:Robust DCDC Converter for Automotive Applications, Analog Circuit Design, Springer Science+Business Media B.V. (2009), pp.269-271.

DOI: 10.1007/978-1-4020-8944-2_15

Google Scholar

[3] L.Y. Zhang:Recurrent Neural Network Optimization for Model Predictive Control,IEEE(2008).

Google Scholar

[4] J.K. Gruber, C.Bordons, F. Dorado: Proceedings of the American Control Conference, Seattle, USA, June 11–13 (2008).

Google Scholar

[5] Y.Xia, G.Feng, J.Wang: A Recurrent neural network with exponential convergence for solving conves quadratic program and linear piecewise equation, Neural Network. Vol.17 (2004), pp.1003-1015.

DOI: 10.1016/j.neunet.2004.05.006

Google Scholar

[6] M.S. Bazaraa, H.D. Sherali, C.M. Shetty in:Nonlinear programming—theory and algorithms ( Wiley,New York 1993).

Google Scholar