Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty

By Jeremy A. Templeton. 2015

388 citations recorded by the source corpus.

Subfields: Computational Mechanics, Statistical and Nonlinear Physics, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Statistical and Nonlinear Physics

In Statistics and Probability

Citations in the corpus

Theory-Guided Data Science: A New Paradigm for Scientific Discovery from Data (2016, background): “, scaling, translation, and rotational invariance in turbulence modeling [78], and symmetry invariance in computational chemistry [26], [79].”
Quantification of model uncertainty in RANS simulations: A review (2018, methodology): “ection function (q) for the baseline model obtained by posing the discrepancy term as a function of non-dimensionalized mean ow variables (e.g., S and , both properly normalized with local quantities [120], as well as the ratio P!=D! between production and dissipation [121]). Choosing ow variables q”
Turbulence Modeling in the Age of Data (2018, methodology): “forest predicted markers (blue and maroon) showing regions with possible emerging of negative eddy viscosity in several ows, indicating likely failure of linear eddy viscosity models in such regions (Ling and Templeton 2015). 5.Predictive modeling using data-driven techniques The eorts discussed in ”
Physics-informed machine learning approach for augmenting turbulence models: A comprehensive framework (2017, methodology): “To ensure non-dimensionality of the raw inputs, the normalization scheme used in [8] is adopted.”
Physics-informed machine learning approach for reconstructing Reynolds stress modeling discrepancies based on DNS data (2016, methodology): “y in three-dimensional ows, but the test cases presented here are two-dimensional ow. We excluded two additional features related to linear and nonlinear eddy viscosities (features 7 6 and 12 in ref. [12]). These quantities were specically chosen for evaluating qualitative condence indicators of RAN”
Tackling Climate Change with Machine Learning (2019, background): “rther future, researchers are developing ML approaches for a wide range of scientiﬁc modeling challenges, including crash prediction [509], adaptive numerical meshing [510], uncertainty quantiﬁcation [511, 512] and performance optimization [513]. If these strategies are effective, they may solve som”
Machine learning methods for turbulence modeling in subsonic flows around airfoils (2018, background): “lassic RANS models, which further extended the application of machine learning to turbulence modeling and verified the positive prospect of data-driven methods. For example, Ling and Templeton et al. [20, 21, 22, 23] embedded the invariance property into deep neural networks firstly and demonstrated”