Comparison between PSO and OLS for NARX parameter estimation of a DC motor

Author

Mohamad, M.S.A. ; Yassin, I.M. ; Zabidi, Azlee ; Taib, M.N. ; Adnan, R.

Author_Institution

Fac. of Electr. Eng., Univ. Teknol. Mara, Shah Alam, Malaysia

fYear

2013

fDate

22-25 Sept. 2013

Firstpage

27

Lastpage

32

Abstract

Recent works suggest that the Particle Swarm Optimization (PSO) algorithm is a highly-efficient optimization technique for structure selection of NARMAX and its derivative models. This research extends those findings by proposing PSO for parameter estimation of a Nonlinear Auto-Regressive with Exogenous (NARX) model for a Direct Current (DC) motor. The proposed method was compared to the established Orthogonal Least Squares (OLS) method. The findings indicate that PSO was comparable to OLS in solving the Least Squares (LS) parameter estimation problem posed in the NARX model.

Keywords

DC motors; autoregressive processes; parameter estimation; particle swarm optimisation; regression analysis; DC motor; NARMAX structure selection; NARX parameter estimation; OLS; PSO; derivative models; direct current motor; nonlinear autoregressive-with-exogenous model; orthogonal least squares method; particle swarm optimization algorithm; DC motors; Industrial electronics; Mathematical model; Optimization; Parameter estimation; Silicon; Training; DC Motor; NARX; Nonlinear System Identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics and Applications (ISIEA), 2013 IEEE Symposium on

Conference_Location

Kuching

Print_ISBN

978-1-4799-1124-0

Type

conf

DOI

10.1109/ISIEA.2013.6738962

Filename

6738962