Parametric complexity reduction of the Meixner-like model using genetic algorithms

Maraoui, Safa; Bouzrara, Kais

doi:10.1109/cadiag.2017.8075705

Cited by 5 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The optimization problem of the pole in the Meixner-like model was addressed by Maraoui et al (2016a, 2016b) using gradient and Newton-Raphson type iterative algorithms that deal with black box context where only the response of the system to a persistently exciting input signal is known. Also, a method based on Genetic algorithms is proposed, in Maraoui and Bouzrara (2017), to optimize the Meixner-like pole.…”

Section: Optimization Of the Poles And The Generalization Orders Of Tmentioning

confidence: 99%

“…To ensure the reduction of the model coefficient number, the parameters characterizing the Meixner-like functions have to be optimized. To do so, we propose to optimize the Meixner-like poles by using Genetic algorithms that are widely adopted in recent years and that are very powerful in stochastic system modeling (Akbari and Ziarati, 2011; Maraoui and Bouzrara, 2017; Zhang et al, 2007). The processing of these algorithms can converge to the global optimum with a high probability.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Parametric identification of linear systems using AutoRegressive with eXogenous input model expansion on Meixner-like orthonormal bases

Maraoui

Bouzrara

2019

Transactions of the Institute of Measurement and Control

Self Cite

View full text Add to dashboard Cite

This paper proposes a new model for a representation of linear system with delay. This latter, is given by decomposing the coefficients of the AutoRegressive with eXogenous input (ARX) model associated with the output and the input on two independent Meixner-like orthonormal bases. The resulting model is called ARX Meixner-like model. In order to ensure the parameter complexity reduction, the Meixner-like poles are optimized using Genetic algorithms method. Unknown But Bounded Error (UBBE) approaches are used to update the uncertainty domain of the parameters of the new obtained model. A numerical example of system with delay and two examples of experimental research are made to prove the efficiency of the proposed approach.

show abstract

Section: Optimization Of the Poles And The Generalization Orders Of Tmentioning

confidence: 99%