A neural-network-augmented differential equation (UDE) reconstructs the missing constitutive terms of UCM, Johnson-Segalman, and Giesekus models from shear-stress data alone, predicts unobserved normal stresses, but only approximates the exponential PTT model.
Nonlinear rheology of colloidal dispersions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.flu-dyn 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Finding the Underlying Viscoelastic Constitutive Equation via Universal Differential Equations and Differentiable Physics
A neural-network-augmented differential equation (UDE) reconstructs the missing constitutive terms of UCM, Johnson-Segalman, and Giesekus models from shear-stress data alone, predicts unobserved normal stresses, but only approximates the exponential PTT model.