Title: A novel method for optimal model simplification of large scale linear discrete-time systems

Authors: Ganji Vasu; Mangipudi Sivakumar; Manyala Ramalinga Raju

Addresses: Department of Electrical and Electronics Engineering, J.N.T. University Kakinada, Kakinada-533003, Andhra Pradesh, India ' Department of Electrical and Electronics Engineering, Gudlavalleru Engineering College, Gudlavalleru-521356, Krishna, Andhra Pradesh, India ' Department of Electrical and Electronics Engineering, J.N.T. University Kakinada, Kakinada-533003, Andhra Pradesh, India

Abstract: An accurate and stability preserving model order reduction method for large scale linear discrete-time systems is presented in this paper. The denominator coefficients of the reduced order model are obtained by least squares matching of time-moment proportional's and Markov parameters of the original system and reduced order model about (z − 1), while the numerator coefficients are determined by minimising the integral square error (ISE) between the unit step responses of the original system and its reduced order approximate. Instead of evaluating ISE from time responses of the system and its approximate, a matrix formula is developed in this paper. Three examples are provided to show the efficacy of the proposed method.

Keywords: model order reduction; time moment proportionals; Markov parameters; least squares matching; discrete integral square errors; optimisation; model simplification; large scale systems; linear discrete-time systems.

DOI: 10.1504/IJAAC.2016.076455

International Journal of Automation and Control, 2016 Vol.10 No.2, pp.120 - 141

Received: 25 May 2015
Accepted: 13 Jan 2016
Published online: 09 May 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article