Horizon Selection in Physics-Enhanced Neural ODEs: Theoretical Insights and Flux Linkage Application
arXiv:2607.25804
The paper investigates how the integration horizon used during training influences physics-enhanced Neural ODEs, proposing longer horizons to reduce bias in physical parameter estimation and improve generalization, and demonstrates the approach on flux map and resistance estimation for a permanent magnet synchronous machine.
Abstract
The integration horizon during the training plays a critical role in Physics-Enhanced Neural Ordinary Differential Equations. We draw conclusions about horizon extension in the training of Neural Ordinary Differential Equations based on classical nonlinear system identification of input-output models. In light of this insight, we propose a framework that exploits longer horizons to reduce bias in physical parameter estimates, extracts residual information from data, and acts as a regularizer improving generalization. In the learning of a model for permanent magnet synchronous machine, the method is used to jointly estimate the flux map and the resistance.
Accepted for presentation at IFAC World Congress 2026