Alert button

Neural network based generation of 1-dimensional stochastic fields with turbulent velocity statistics

Nov 21, 2022
Carlos Granero-Belinchon

Figure 1 for Neural network based generation of 1-dimensional stochastic fields with turbulent velocity statistics
Figure 2 for Neural network based generation of 1-dimensional stochastic fields with turbulent velocity statistics
Figure 3 for Neural network based generation of 1-dimensional stochastic fields with turbulent velocity statistics
Figure 4 for Neural network based generation of 1-dimensional stochastic fields with turbulent velocity statistics

Share this with someone who'll enjoy it:

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.

View paper onarxiv icon

Share this with someone who'll enjoy it: