A recursive neural-network-based subgrid-scale model for large eddy simulation: application to homogeneous isotropic turbulence

HIGHLIGHTS

• What: The aim of this study is to devise a new NN-based SGS model designed to overcome the limitations of existing direct NN-based SGS models. The authors develop an NN-based SGS model whose input and output are Δ̄2 |ᾱ|ᾱij and τijr, respectively (NN-based velocity gradient model, called NNVGM hereafter). The authors aimed at overcoming the difficulties encountered during the SGS model development, i.e. how to obtain high-Reynolds-number data for training and how to apply a trained NN to untrained flow.
• Who: J. Fluid Mech. and colleagues from the . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.