Nonlinear port-Hamiltonian system identification from input-state-output data

Karim Cherifi, Achraf El Messaoudi, Hannes Gernandt, Marco Roschkowski

correctmedium confidence

Category: math.DS
Journal tier: Specialist/Solid
Processed: Sep 28, 2025, 12:56 AM
arXiv Links: Abstract ↗PDF ↗

Audit review

The paper’s parametrization directly enforces J(x) = −J(x)⊤ and R(x) = R(x)⊤ ≥ 0 “by construction,” and the candidate solution spells out the same reasoning with an explicit reshaping argument and a standard passivity/energy-balance derivation. There are no logical gaps relative to the stated claims; the model’s derivation of dH/dt = −∇H⊤R∇H + y⊤u ≤ y⊤u is standard and consistent with the paper’s framework.

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The paper’s reshape-based parametrization enforces the pH constraints transparently and is implemented in a neural identification setting with reasonable experiments. The algebraic claims checked here are correct. Including a short, explicit energy-balance derivation and clarifying minimal regularity assumptions would slightly improve accessibility, but the contribution is otherwise sound.