Move page script: Move page script moved page Kruz et al 2021 to Kurz et al 2021

2021-04-27T11:09:24Z

Move page script moved page Kruz et al 2021 to Kurz et al 2021

Move page script: Move page script moved page Beck et al 2021a to Kruz et al 2021

2021-04-16T14:28:54Z

Move page script moved page Beck et al 2021a to Kruz et al 2021

Scipediacontent: Scipediacontent moved page Draft Content 451911896 to Beck et al 2021a

2021-03-11T15:42:42Z

Scipediacontent moved page Draft Content 451911896 to Beck et al 2021a

Scipediacontent: Created page with "== Abstract == In the present work, we investigate the stability of turbulence closure predictions from neural network models and highlight the role of model-data-inconsisten..."

2021-03-11T15:42:40Z

Created page with "== Abstract == In the present work, we investigate the stability of turbulence closure predictions from neural network models and highlight the role of model-data-inconsisten..."

New page

== Abstract ==

In the present work, we investigate the stability of turbulence closure predictions from neural network models and highlight the role of model-data-inconsistency during inference. We quantify this inconsistency by applying the Mahalanobis distance and demonstrate that the instability of the model predictions in practical large eddy simulations (LES) correlates with a deviation of the input data between the training dataset and actual simulation data. Moreover, the method of 'stability training' is applied to increase the robustness of recurrent artificial neural networks (ANN) against small perturbations in the input, which are typically unavoidable in any practical scenario. We show that this method can increase the stability of simulations with ANN-based closure term predictions significantly. The models also achieve good accuracy on the blind testing set in comparison to the baseline model trained without stability training. The work presented here can thus be seen as a building block towards long-term stable data-driven models for dynamical systems and highlights methods to detect and counter model-data-inconsistencies.

== Full document ==
<pdf>Media:Draft_Content_451911896p1012.pdf</pdf>

← Older revision	Revision as of 11:09, 27 April 2021
(No difference)

← Older revision	Revision as of 14:28, 16 April 2021
(No difference)

← Older revision	Revision as of 15:42, 11 March 2021
(No difference)

Kurz et al 2021 - Revision history

Move page script: Move page script moved page Kruz et al 2021 to Kurz et al 2021

Move page script: Move page script moved page Beck et al 2021a to Kruz et al 2021

Scipediacontent: Scipediacontent moved page Draft Content 451911896 to Beck et al 2021a

Scipediacontent: Created page with "== Abstract == In the present work, we investigate the stability of turbulence closure predictions from neural network models and highlight the role of model-data-inconsisten..."