Real-time identified chaotic plants using neural enhanced learning machine technique

Purpose This paper aims to propose a new neural-based enhanced extreme learning machine (EELM) algorithm, used as an online adaptive estimation model, regarding undetermined system dynamics and containing internal/external perturbations. Design/methodology/approach The EELM structure bases on the si...

Full description

Saved in:

Bibliographic Details
Published in:	Engineering computations 2021-07, Vol.38 (6), p.2810-2832
Main Author:	Anh, Ho Pham Huy
Format:	Article
Language:	eng
Subjects:	Algorithms Artificial neural networks Chaos theory Deep learning Fault diagnosis Identification Machine learning Neural networks Perturbation System dynamics Systems stability Weighting
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Purpose This paper aims to propose a new neural-based enhanced extreme learning machine (EELM) algorithm, used as an online adaptive estimation model, regarding undetermined system dynamics and containing internal/external perturbations. Design/methodology/approach The EELM structure bases on the single layer feed-forward neural (SLFN) model in which the hidden weighting coefficients are initiated in random and the weighting outputs of the SLFN are online modified using an online adaptive rule implemented from Lyapunov stability concept. Findings Four different benchmark uncertain chaotic system tests have been satisfactorily investigated for demonstrating the superiority of proposed EELM technique. Originality/value Authors confirm that this manuscript is original.
ISSN:	0264-4401 1758-7077