Andrea Beck and her research assistant Marius Kurz from PN 1-1 have just published an Open Access review article on machine learning in the “GAMM Mitteilungen”. The purpose of the article is to organize the many different machine Learning applications in the field of turbulence modelling and to think about general problems and challenges that need to be solved to make ML successful in that field of research. Rather than focussing on a specific method, the article discusses how ML-driven approaches and classical approximation methods can complement and interact with each other.
“The idea of the article is to give “beginners” in Machine Learning, who come from the fluid mechanics community, a compact overview of what is currently possible, what is still lacking and which general problems have to be solved”, explains Andrea Beck the motivation for the article.
Abstract: This work presents a review of the current state of research in data‐driven turbulence closure modeling. It offers a perspective on the challenges and open issues but also on the advantages and promises of machine learning (ML) methods applied to parameter estimation, model identification, closure term reconstruction, and beyond, mostly from the perspective of large Eddy simulation and related techniques. We stress that consistency of the training data, the model, the underlying physics, and the discretization is a key issue that needs to be considered for a successful ML‐augmented modeling strategy. In order to make the discussion useful for non‐experts in either field, we introduce both the modeling problem in turbulence as well as the prominent ML paradigms and methods in a concise and self‐consistent manner. In this study, we present a survey of the current data‐driven model concepts and methods, highlight important developments, and put them into the context of the discussed challenges.
Authors: Andrea Beck, Marius Kurz
Read the full article here https://onlinelibrary.wiley.com/doi/10.1002/gamm.202100002.