Sub-grid scale model classification and blending through deep learning

TLDR

The study applies machine learning to classify and blend turbulence models for large‑eddy simulations of turbulent flows. A predictive framework computes local conditional probabilities to assign each grid point to a functional, structural, or no‑model hypothesis and blends models accordingly, demonstrated on Kraichnan turbulence. The approach yields robust, stable closures and can combine strengths of different models for complex flow predictions.

Abstract

In this article we detail the use of machine learning for spatiotemporally dynamic turbulence model classification and hybridization for the large eddy simulations (LES) of turbulence. Our predictive framework is devised around the determination of local conditional probabilities for turbulence models that have varying underlying hypotheses. As a first deployment of this learning, we classify a point on our computational grid as that which requires the functional hypothesis, the structural hypothesis or no modeling at all. This ensures that the appropriate model is specified from \emph{a priori} knowledge and an efficient balance of model characteristics is obtained in a particular flow computation. In addition, we also utilize the conditional probability predictions of the same machine learning to blend turbulence models for another hybrid closure. Our test-case for the demonstration of this concept is given by Kraichnan turbulence which exhibits a strong interplay of enstrophy and energy cascades in the wave number domain. Our results indicate that the proposed methods lead to robust and stable closure and may potentially be used to combine the strengths of various models for complex flow phenomena prediction.