EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics-Informed Neural Networks for Higher-Order Nonlinear Schrödinger Equations: Soliton Dynamics in External Potentials

Leonid Serkin, Tatyana L. Belyaeva

Mathematics · 04.06.2025 · 13 (11) · 1882

DOI: 10.3390/math13111882 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mathematics, online erschienen am 04.06.2025. Verfasst von Leonid Serkin und 1 weiteren Autor:innen. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 149 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This review summarizes the application of physics-informed neural networks (PINNs) for solving higher-order nonlinear partial differential equations belonging to the nonlinear Schrödinger equation (NLSE) hierarchy, including models with external potentials. We analyze recent studies in which PINNs have been employed to solve NLSE-type evolution equations up to the fifth order, demonstrating their ability to obtain one- and two-soliton solutions, as well as other solitary waves with high accuracy. To provide benchmark solutions for training PINNs, we employ analytical methods such as the nonisospectral generalization of the AKNS scheme of the inverse scattering transform and the auto-Bäcklund transformation. Finally, we discuss recent advancements in PINN methodology, including improvements in network architecture and optimization techniques.

Zitieren

APA

Serkin, L., & Belyaeva, T. L. (2025). Physics-Informed Neural Networks for Higher-Order Nonlinear Schrödinger Equations: Soliton Dynamics in External Potentials. Mathematics, 13(11), 1882. https://doi.org/10.3390/math13111882

Vancouver

Serkin L, Belyaeva TL. Physics-Informed Neural Networks for Higher-Order Nonlinear Schrödinger Equations: Soliton Dynamics in External Potentials. Mathematics. 2025;13(11):1882. doi:10.3390/math13111882

BibTeX
@article{Serkin2025,
  author = {Serkin, Leonid and Belyaeva, Tatyana L.},
  title = {Physics-Informed Neural Networks for Higher-Order Nonlinear Schrödinger Equations: Soliton Dynamics in External Potentials},
  journal = {Mathematics},
  year = {2025},
  volume = {13},
  number = {11},
  pages = {1882},
  doi = {10.3390/math13111882}
}
Kostenlos starten →