Identification of structured nonlinear state–space models for hysteretic systems using neural network hysteresis operators

Konstantinos Krikelis, Jin Song Pei, Koos van Berkel, Maarten Schoukens (Corresponding author)

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

51 Downloads (Pure)

Samenvatting

Hysteretic system behavior is ubiquitous in science and engineering fields including measurement systems and applications. In this paper, we put forth a nonlinear state–space system identification method that combines the state–space equations to capture the system dynamics with a compact and exact artificial neural network (ANN) representation of the classical Prandtl–Ishlinskii (PI) hysteresis. These ANN representations called PI hysteresis operator neurons employ recurrent ANNs with classical activation functions, and thus can be trained with classical neural network learning algorithms. The structured nonlinear state–space model class proposed in this paper, for the first time, offers a flexible interconnection of PI hysteresis operators with a linear state–space model through a linear fractional representation. This results in a comprehensive and flexible model structure. The performance is validated both on numerical simulation and on measurement data.

Originele taal-2Engels
Artikelnummer113966
Aantal pagina's12
TijdschriftMeasurement: Journal of the International Measurement Confederation
Volume224
DOI's
StatusGepubliceerd - jan. 2024

Bibliografische nota

Publisher Copyright:
© 2023 The Authors

Vingerafdruk

Duik in de onderzoeksthema's van 'Identification of structured nonlinear state–space models for hysteretic systems using neural network hysteresis operators'. Samen vormen ze een unieke vingerafdruk.

Citeer dit