ST-PCNN: Spatio-Temporal Physics-Coupled Neural Networks for Dynamics Forecasting

TL;DR

The paper introduces ST-PCNN, a neural network model that learns physics parameters and leverages spatio-temporal correlations to improve long-range forecasting of dynamical systems like ocean currents.

Contribution

It proposes a novel physics-coupled neural network that learns physics parameters and models local spatio-temporal correlations for better dynamics forecasting.

Findings

Outperforms existing physics-informed models on ocean current data

Achieves accurate long-range forecasts exceeding 30 steps

Effectively learns underlying physics parameters from data

Abstract

Ocean current, fluid mechanics, and many other spatio-temporal physical dynamical systems are essential components of the universe. One key characteristic of such systems is that certain physics laws -- represented as ordinary/partial differential equations (ODEs/PDEs) -- largely dominate the whole process, irrespective of time or location. Physics-informed learning has recently emerged to learn physics for accurate prediction, but they often lack a mechanism to leverage localized spatial and temporal correlation or rely on hard-coded physics parameters. In this paper, we advocate a physics-coupled neural network model to learn parameters governing the physics of the system, and further couple the learned physics to assist the learning of recurring dynamics. A spatio-temporal physics-coupled neural network (ST-PCNN) model is proposed to achieve three goals: (1) learning the underlying physics parameters, (2) transition of local information between spatio-temporal regions, and (3) forecasting future values for the dynamical system. The physics-coupled learning ensures that the proposed model can be tremendously improved by using learned physics parameters, and can achieve good long-range forecasting (e.g., more than 30-steps). Experiments, using simulated and field-collected ocean current data, validate that ST-PCNN outperforms existing physics-informed models.