A PINN that minimizes Stokes-equation residuals recovers Herschel-Bulkley, Carreau and Papanastasiou parameters from noisy synthetic velocity data, with an AIC-style loss for model selection.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.flu-dyn 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Inferring viscoplastic models from velocity fields: a physics-informed neural network approach
A PINN that minimizes Stokes-equation residuals recovers Herschel-Bulkley, Carreau and Papanastasiou parameters from noisy synthetic velocity data, with an AIC-style loss for model selection.