Physics-informed neural networks with Riccati-type pseudopotential constraints for solving nonlinear evolution equations
Chaos: An Interdisciplinary Journal of Nonlinear Science · 01.09.2026 · 36 (9)
DOI: 10.1063/5.0348469 ↗Kurz gefasst
Fachartikel in der Zeitschrift Chaos: An Interdisciplinary Journal of Nonlinear Science, online erschienen am 23.09.2026. Verfasst von Junchao Wan und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Science, Shanghai Maritime University, Shanghai 201306, People's Republic of China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 56 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Physics-informed neural networks (PINNs) provide a powerful framework for solving nonlinear evolution equations, but their accuracy and stability often deteriorate for systems with high-frequency structures, strong nonlinear interactions, or high-order derivatives. Although structure-informed extensions can improve physical consistency by incorporating geometric or integrable constraints, many such formulations rely on high-dimensional auxiliary variables, which may introduce latent gauge redundancies and enlarge the optimization space. To circumvent these inherent limitations, we propose RP-PINN, a physics-informed architecture based on Riccati-type pseudopotentials. By combining the Wahlquist–Estabrook prolongation structure with Nucci’s direct method, the proposed framework replaces high-dimensional auxiliary representations with Riccati-type equations, which are embedded into the loss function as low-order rigid constraints. Numerical experiments on the Korteweg–de Vries, nonlinear Schrödinger, sine-Gordon, and Harry Dym equations demonstrate that RP-PINN can accurately reproduce soliton, rogue wave, and kink solutions while mitigating the adverse effects of high-order automatic differentiation. These results demonstrate the effectiveness and applicability of RP-PINN to integrable systems admitting Riccati-type pseudopotentials.
Autor:innen und Institutionen
- School of Science, Shanghai Maritime University, Shanghai 201306, People's Republic of China.
- College of Mathematics and Systems Science, Shandong University of Science and Technology, Qingdao 266590, People's Republic of China.
ORCID-Profile: Junchao Wan, Hui Wang, Yunhu Wang
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Wan, J., Wang, H., & Wang, Y. (2026). Physics-informed neural networks with Riccati-type pseudopotential constraints for solving nonlinear evolution equations. Chaos: An Interdisciplinary Journal of Nonlinear Science, 36(9). https://doi.org/10.1063/5.0348469
Vancouver
Wan J, Wang H, Wang Y. Physics-informed neural networks with Riccati-type pseudopotential constraints for solving nonlinear evolution equations. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2026;36(9). doi:10.1063/5.0348469
BibTeX
@article{Wan2026,
author = {Wan, Junchao and Wang, Hui and Wang, Yunhu},
title = {Physics-informed neural networks with Riccati-type pseudopotential constraints for solving nonlinear evolution equations},
journal = {Chaos: An Interdisciplinary Journal of Nonlinear Science},
year = {2026},
volume = {36},
number = {9},
doi = {10.1063/5.0348469}
}
EUVIMED