Bounded nonlinear forecasts of partially observed geophysical systems with physics-constrained deep learning
Type | Article | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Date | 2023-04 | ||||||||||||
Language | English | ||||||||||||
Author(s) | Ouala Said1, Brunton Steven L.2, Chapron Bertrand3, Pascual Ananda4, Collard Fabrice5, Gaultier Lucile5, Fablet Ronan1 | ||||||||||||
Affiliation(s) | 1 : IMT Atlantique; Lab-STICC, 29200 Brest, France 2 : University of Washington, USA 3 : Ifremer, Lops, 29200 Brest, France 4 : IMEDEA, UIB-CSIC, 07190 Esporles, Spain 5 : ODL, 29200 Brest, France |
||||||||||||
Source | Physica D-nonlinear Phenomena (0167-2789) (Elsevier BV), 2023-04 , Vol. 446 , P. 133630 (19p.) | ||||||||||||
DOI | 10.1016/j.physd.2022.133630 | ||||||||||||
WOS© Times Cited | 5 | ||||||||||||
Keyword(s) | Partially-observed systems, Embedding, Boundedness, Deep learning, Neural ODE, Forecasting | ||||||||||||
Abstract | The complexity of real-world geophysical systems is often compounded by the fact that the observed measurements depend on hidden variables. These latent variables include unresolved small scales and/or rapidly evolving processes, partially observed couplings, or forcings in coupled systems. This is the case in ocean-atmosphere dynamics, for which unknown interior dynamics can affect surface observations. The identification of computationally-relevant representations of such partially-observed and highly nonlinear systems is thus challenging and often limited to short-term forecast applications. Here, we investigate the physics-constrained learning of implicit dynamical embeddings, leveraging neural ordinary differential equation (NODE) representations. In particular, we restrict the NODE representation to linear-quadratic dynamics and enforce a global boundedness constraint, which promotes the generalization of the learned dynamics to arbitrary initial conditions. The proposed architecture is implemented within a deep learning framework, and its relevance is demonstrated with respect to state-of-the-art schemes for different case studies representative of geophysical dynamics. |
||||||||||||
Full Text |
|