KFU. Card publication. Physics-Informed Neural Networks for Limit-Cycle Analysis in Nonlinear Dynamical Systems with Absolute-Value Nonlinearit

- RUS
- 中文
- ES

PHYSICS-INFORMED NEURAL NETWORKS FOR LIMIT-CYCLE ANALYSIS IN NONLINEAR DYNAMICAL SYSTEMS WITH ABSOLUTE-VALUE NONLINEARIT

Form of presentation

Conference proceedings in Russian journals and collections

Year of publication

2025

��

��

Authors, employees of KFU

Arabov Mullosharaf Kurbonovich, author

Other authors

- - -, author

Bibliographic description in the original language

Arabov M. K. Physics-Informed Neural Networks for Limit-Cycle Analysis in Nonlinear Dynamical Systems with Absolute-Value Nonlinearity // Materialy mezhdunarodnoy nauchnoy konferencii «Ufimskaya osennyaya matematicheskaya shkola» (g. Ufa, 1–5 oktyabrya 2025 g.). V 2 t. / otv. red. Z. Yu. Fazullin. — Ufa: Aeterna, 2025. — T. 2. — S. 102–104.

Annotation

�� «�� »

Keywords

physics-informed neural networks, nonlinear dynamical systems, limit cycles, absolute-value nonlinearity

The name of the journal

�� «�� »

Please use this ID to quote from or refer to the card

https://repository.kpfu.ru/eng/?p_id=317850&p_lang=2

Resource files

File name	Size (MB)	Format
F_UOMSh_2025.pdf	0,31	pdf	show / download

Full metadata record

Field DC	Value	Language
dc.contributor.author	Arabov Mullosharaf Kurbonovich	ru_RU
dc.contributor.author	- - -	ru_RU
dc.date.accessioned	2025-01-01T00:00:00Z	ru_RU
dc.date.available	2025-01-01T00:00:00Z	ru_RU
dc.date.issued	2025	ru_RU
dc.identifier.citation	�� . �. Physics-Informed Neural Networks for Limit-Cycle Analysis in Nonlinear Dynamical Systems with Absolute-Value Nonlinearity // �� «�� » (�. ��, 1–5 �� 2025 �.). � 2 �. / ��. ��. �. �. ��. — ��: ��, 2025. — �. 2. — �. 102–104.	ru_RU
dc.identifier.uri	https://repository.kpfu.ru/eng/?p_id=317850&p_lang=2	ru_RU
dc.description.abstract	�� «�� »	ru_RU
dc.description.abstract	We apply Physics-Informed Neural Networks (PINNs) to second-order nonlinear dynamical systems with an absolute-value nonlinearity. The focus is on detecting and quantitatively characterising limit cycles. The PINN is trained using the L-BFGS quasi-Newton optimiser. Results are validated against numerical solvers and further analysed via Fourier spectral analysis (FFT) of the time series, enabling the amplitude and frequency characteristics of the cycles to be assessed. Keywords: physics-informed neural networks, nonlinear dynamical systems, limit cycles, absolute-value nonlinearity	ru_RU
dc.language.iso	ru	ru_RU
dc.subject	physics-informed neural networks	ru_RU
dc.subject	nonlinear dynamical systems	ru_RU
dc.subject	limit cycles	ru_RU
dc.subject	absolute-value nonlinearity	ru_RU
dc.title	Physics-Informed Neural Networks for Limit-Cycle Analysis in Nonlinear Dynamical Systems with Absolute-Value Nonlinearit	ru_RU
dc.type	Conference proceedings in Russian journals and collections	ru_RU