A conditional GAN with a built-in dynamics block can predict parameterized fluid flows, with accuracy that degrades at high Reynolds numbers and a sweet spot in the number of training steps.
Computers & fluids 35 (3), 326--348
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Data-driven Modeling of Parameterized Nonlinear Fluid Dynamical Systems with a Dynamics-embedded Conditional Generative Adversarial Network
A conditional GAN with a built-in dynamics block can predict parameterized fluid flows, with accuracy that degrades at high Reynolds numbers and a sweet spot in the number of training steps.