SPIN-ODE: Stiff Physics-Informed Neural ODE for Chemical Reaction Rate Estimation

TL;DR

This paper introduces SPIN-ODE, a novel neural ODE framework designed to accurately estimate chemical reaction rate coefficients in stiff systems, overcoming training challenges and improving robustness in complex atmospheric chemistry modeling.

Contribution

The paper presents the first neural ODE approach specifically tailored for stiff chemical systems, with a three-stage optimization process for improved rate coefficient estimation.

Findings

Effective in both synthetic and real-world datasets

Addresses training instability in stiff systems

Enhances chemical reaction modeling accuracy

Abstract

Estimating rate coefficients from complex chemical reactions is essential for advancing detailed chemistry. However, the stiffness inherent in real-world atmospheric chemistry systems poses severe challenges, leading to training instability and poor convergence, which hinder effective rate coefficient estimation using learning-based approaches. To address this, we propose a Stiff Physics-Informed Neural ODE framework (SPIN-ODE) for chemical reaction modelling. Our method introduces a three-stage optimisation process: first, a black-box neural ODE is trained to fit concentration trajectories; second, a Chemical Reaction Neural Network (CRNN) is pre-trained to learn the mapping between concentrations and their time derivatives; and third, the rate coefficients are fine-tuned by integrating with the pre-trained CRNN. Extensive experiments on both synthetic and newly proposed real-world datasets validate the effectiveness and robustness of our approach. As the first work addressing stiff neural ODE for chemical rate coefficient discovery, our study opens promising directions for integrating neural networks with detailed chemistry.