Data-driven discovery of conservation laws from trajectories via neural deflation

Open Webpage

Shaoxuan Chen, Panayotis G. Kevrekidis, Hong-Kun Zhang, Wei Zhu

Published: 2025, Last Modified: 15 May 2025Commun. Nonlinear Sci. Numer. Simul. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Data-driven discovery of conservation laws in nonlinear dynamical systems.•Identification of a complete set of functionally independent, Poisson-commuting conservation laws.•Method developed directly from system trajectories, without explicit knowledge of underlying equations of motion.•Examples include 1D and 2D harmonic oscillators, the Toda lattice, the Fermi–Pasta–Ulam–Tsingou lattice, and the Calogero-Moser system.