Learning Groundwater Contaminant Diffusion‐Sorption Processes With a Finite Volume Neural Network. Issue 12 (29th November 2022)

• Record Type: Journal Article
• Authors: Praditia, Timothy; Karlbauer, Matthias; Otte, Sebastian; Oladyshkin, Sergey; Butz, Martin V.; Nowak, Wolfgang
• Is Part Of: Water resources research. Volume 58:Issue 12(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-11-29
• Subjects: artificial neural networks -- hybrid model -- physics‐informed neural networks -- machine learning -- diffusion‐sorption -- groundwater contamination; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2022WR033149
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Improved understanding of complex hydrosystem processes is key to advance water resources research. Nevertheless, the conventional way of modeling these processes suffers from a high conceptual uncertainty, due to almost ubiquitous simplifying assumptions used in model parameterizations/closures. Machine learning (ML) models are considered as a potential alternative, but their generalization abilities remain limited. For example, they normally fail to predict accurately across different boundary conditions. Moreover, as a black box, they do not add to our process understanding or to discover improved parameterizations/closures. To tackle this issue, we propose the hybrid modeling framework FINN (finite volume neural network). It merges existing numerical methods for partial differential equations (PDEs) with the learning abilities of artificial neural networks (ANNs). FINN is applied on discrete control volumes and learns components of the investigated system equations, such as numerical stencils, model parameters, and arbitrary closure/constitutive relations. Consequently, FINN yields highly interpretable results. We demonstrate FINN's potential on a diffusion‐sorption problem in clay. Results on numerically generated data show that FINN outperforms other ML models when tested under modified boundary conditions, and that it can successfully differentiate between the usual, known sorption isotherms. Moreover, we also equip FINN with uncertainty quantification methods to lay open the total uncertainty of scientific learning, and then apply it to a laboratory experiment. The results show that FINN performs better than calibrated PDE‐based models as it is able to flexibly learn and model sorption isotherms without being restricted to choose among available parametric models. Key Points: We propose a hybrid modeling framework combining uncertainty‐quantifying artificial neural networks with physical domain knowledge Our method learns diffusion processes and discovers unknown sorption isotherms, while incorporating different boundary conditions Excellent generalization is retained even when trained with sparse and noisy real‐world data from a laboratory experiment