Influence of State Space Topology on the Parameter Identification Based on the PSO Method

Authors

  • Michal Dub Department of Aircraft Electrical Systems, University of Defence, Brno, Czech Republic
  • Alexandr Stefek Department of Air Defence Systems, University of Defence, Brno, Czech Republic

Keywords:

Parameter estimation, particle swarm optimization, system identification

Abstract

Motion control of electromechanical systems still plays very important role in wide area of weapon systems. Modern control systems use not only data from various sensors but also state parameters of controlled system. The article explores influence of state space topology on parameter identification of real simple electromechanical system based on the Particle Swarm Optimization (PSO) method. Four different but equivalent mathematical models of the second order were used to create different state spaces of the system parameters. A general recommendation for the PSO method setup and two independent program tools were used to evaluate the state space searching by the PSO method. The PSO simulations were focused on both narrow and wide state spaces around the fitness function global minimum. New approach to set the PSO method initial agents’ positions has been introduced because traditional random uniform distribution failed if wide state spaces were used.

Author Biographies

  • Michal Dub, Department of Aircraft Electrical Systems, University of Defence, Brno, Czech Republic

    Department of Aircraft Electrical Systems

  • Alexandr Stefek, Department of Air Defence Systems, University of Defence, Brno, Czech Republic

    Department of Air Defence Systems

References

KENNEDY, J. and EBERHART, R. Particle Swarm Optimization. In Proceedings of IEEE International Conference on Neural Networks. New York: IEEE, 1995, p. 1942-1948, ISBN 0-7803-2769-1.

KENNEDY, J. and EBERHART, R. Swarm Intelligence. San Francisco: Morgan Kaufmann, 2001, ISBN 1-55860-595-9.

DENG, X. System Identification Based on Particle Swarm Optimization Algorithm. In Proceedings of 2009 International Conference on Computational Intelligence and Security. Washington: IEEE, 2009, p. 259-263, 2009, ISBN 978-0-7695-3931-7.

DAI, Y., LIU, L. and SONG J. Complex Nonlinear System Identification Based On Cellular Particle Swarm Optimization. In Proceedings of 2013 IEEE International Conference on Mechatronics and Automation. IEEE, p. 1486-1491, 2013, ISBN 978-1-4673-5560-5.

CHEN, S., MEI, T., LUO, M. and YANG, X. Identification of Nonlinear System Based on a New Hybrid Gradient-Based PSO Algorithm. In Proceedings of 2007 International Conference on Information Acquisition. IEEE, p. 265-268, 2007, ISBN 978-1-4244-1219-8.

KINCL, Z. and KOLKA, Z. Test Frequency Selection Using Particle Swarm Optimization. In Advances in Electrical and Electronic Engineering, 2013, vol. 11, no. 3, p. 507-513, ISSN 1804-3119.

KUO, B. C. Automatic control systems. 7th ed. Prentice-Hall, USA, 1995. ISBN 0-13-304759-8.

PHILLIPS, Ch. L. and HARBOR, R. D. Feedback Control Systems. Prentice-Hall, USA, 1996, ISBN 0-13-371691-0.

STEFEK, Alexandr. Benchmarking of heuristic optimization methods. In Proceedings of 14th International Conference on Mechatronics MECHATRONIKA 2011. Trencin: Alexander Dubcek University of Trencin, 2011, p. 68-71, ISBN 978-808075477-8.

STEFEK, A. Distributed Optimization – Concepts, Ideas and Solutions. In Croatian Journal of Education, 2012, vol. 14, Issue SPECIAL.ISS, p. 161-167, ISSN 1848-5189.

DUB, M. and STEFEK, A. Evaluation of PSO Method Application to DC Machine Experimental Identification. In Proceedings of the International Conference on Military Technology. Brno: University of Defence, 2013, p. 887-892, ISBN 978-80-7231-917-6.

DUB, Michal and STEFEK, Alexandr. Using PSO method for System Identification. In Mechatronics 2013. Recent Technological and Scientific Advances. New York: Springer, 2013, p. 143-150, ISBN 978-3-319-02293-2.

Downloads

Published

29-06-2016

Issue

Section

Research Paper

Categories

How to Cite

Dub, M., & Stefek, A. (2016). Influence of State Space Topology on the Parameter Identification Based on the PSO Method. Advances in Military Technology, 11(1), 43-52. https://www.aimt.cz/index.php/aimt/article/view/1101

Similar Articles

1-10 of 163

You may also start an advanced similarity search for this article.

Most read articles by the same author(s)