Proceedings of the 2019 SIAM International Conference on Data Mining: Physics Guided RNNs for Modeling Dynamical Systems: A Case Study in Simulating Lake Temperature Profiles

TitleProceedings of the 2019 SIAM International Conference on Data Mining: Physics Guided RNNs for Modeling Dynamical Systems: A Case Study in Simulating Lake Temperature Profiles
Publication TypeBook
Year of Publication2019
AuthorsJia, Xiaowei, Willard Jared, Karpatne Anuj, Read Jordan, Zwart Jacob, Steinbach Michael, and Kumar Vipin
Series EditorBerger-Wolf, Tanya, and Chawla Nitesh
Number of Pages558 - 566
PublisherSociety for Industrial and Applied Mathematics
CityPhiladelphia, PA
Keywordsphysics-guided recurrent neural network model
Abstract

This paper proposes a physics-guided recurrent neural network model (PGRNN) that combines RNNs and physics-based models to leverage their complementary strengths and improve the modeling of physical processes. Specifically, we show that a PGRNN can improve prediction accuracy over that of physical models, while generating outputs consistent with physical laws, and achieving good generalizability. Standard RNNs, even when producing superior prediction accuracy, often produce physically inconsistent results and lack generalizability. We further enhance this approach by using a pre-training method that leverages the simulated data from a physics-based model to address the scarcity of observed data. Although we present and evaluate this methodology in the context of modeling the dynamics of temperature in lakes, it is applicable more widely to a range of scientific and engineering disciplines where mechanistic (also known as process-based) models are used, e.g., power engineering, climate science, materials science, computational chemistry, and biomedicine.

URLhttps://epubs.siam.org/doi/book/10.1137/1.9781611975673
DOI10.1137/1.9781611975673