.png){kind=icon}
.png){kind=icon}
Return
Neural-network closure modelling for viscoelastic drag-reducing channel flows
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jnnfm.2026.105568.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Machine learning predictions are found for viscoelastic turbulence in the plane channel. • The combined closure term associated with Reynolds and polymeric stresses is used as the target. • Ill-conditioning of RANS-type equations is bypassed. • Newtonian_RANS-to-viscoelastic_DNS, Newtonian_DNS-to-viscoelastic_DNS, and Newtonian_RANS-to-viscoelastic_DNS are the three explored cases. • After the prediction, the closure term is plugged into a solver to find the mean velocity profile.
Keywords:
viscoelastic turbulence
machine learning
closure modelling
drag reduction
channel flows
Journal
J
IF:
2.8
Papers:
153
Citations:
7.6K