Super-resolution and uncertainty estimation from sparse sensors of dynamical physical systems

Frontiers in Water

Water

Adam M. Collins¹  Matthew W. Farthing²  Orie M. Cecil²  Andrew C. Trautz³  Peter Rivera-Casillas⁴  Sourav Dutta⁵ 

[1] Coastal and Hydraulics Laboratory, U.S. Army Engineer Research and Development Center, Duck, NC, United States;Coastal and Hydraulics Laboratory, U.S. Army Engineer Research and Development Center, Vicksburg, MS, United States;Geotechnical and Structures Laboratory, U.S. Army Engineer Research and Development Center, Vicksburg, MS, United States;Information Technology Laboratory, U.S. Army Engineer Research and Development Center, Vicksburg, MS, United States;Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, Austin, TX, United States;

关键词: super-resolution;    machine learning;    uncertainty;    fluid flow;    sea-surface temperature;    soil moisture;   

DOI  :  10.3389/frwa.2023.1137110

 received in 2023-01-04, accepted in 2023-02-23,  发布年份 2023

来源: Frontiers

【 摘 要 】

The goal of this study is to leverage emerging machine learning (ML) techniques to develop a framework for the global reconstruction of system variables from potentially scarce and noisy observations and to explore the epistemic uncertainty of these models. This work demonstrates the utility of exploiting the stochasticity of dropout and batch normalization schemes to infer uncertainty estimates of super-resolved field reconstruction from sparse sensor measurements. A Voronoi tessellation strategy is used to obtain a structured-grid representation from sensor observations, thus enabling the use of fully convolutional neural networks (FCNN) for global field estimation. An ensemble-based approach is developed using Monte-Carlo batch normalization (MCBN) and Monte-Carlo dropout (MCD) methods in order to perform approximate Bayesian inference over the neural network parameters, which facilitates the estimation of the epistemic uncertainty of predicted field values. We demonstrate these capabilities through numerical experiments that include sea-surface temperature, soil moisture, and incompressible near-surface flows over a wide range of parameterized flow configurations.

【 授权许可】

Unknown   
Copyright © 2023 Collins, Rivera-Casillas, Dutta, Cecil, Trautz and Farthing.