Neural network based generation of 1-dimensional stochastic fields with turbulent velocity statistics
Preprint
Abstract
We define and study a fully-convolutional neural network stochastic model, NN-Turb, which generates 1-dimensional fields with turbulent velocity statistics. Thus, the generated process satisfies the Kolmogorov 2/3 law for second order structure function. It also presents negative skewness across scales (i.e. Kolmogorov 4/5 law) and exhibits intermittency. Furthermore, our model is never in contact with turbulent data and only needs the desired statistical behavior of the structure functions across scales for training.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- Deep unsupervised learning of turbulence for inflow generation at various Reynolds numbersJournal of Computational Physics, 2020
- On a Skewed and Multifractal Unidimensional Random Field, as a Probabilistic Representation of Kolmogorov’s Views on TurbulenceAnnales Henri Poincare, 2019
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- A phenomenological theory of Eulerian and Lagrangian velocity fluctuations in turbulent flowsComptes Rendus. Physique, 2012
- Hydrodynamic Turbulence and Intermittent Random FieldsCommunications in Mathematical Physics, 2008
- Towards a complex system approach for the study of turbulenceChemical Engineering Science, 2007
- A New Universal Scaling for Fully Developed Turbulence: The Distribution of Velocity Increments1990
- High-order velocity structure functions in turbulent shear flowsJournal of Fluid Mechanics, 1984
- A refinement of previous hypotheses concerning the local structure of turbulence in a viscous incompressible fluid at high Reynolds numberJournal of Fluid Mechanics, 1962
- Pressure fluctuations in isotropic turbulenceMathematical Proceedings of the Cambridge Philosophical Society, 1951