Deep Learning Convective Flow Using Conditional Generative Adversarial Networks

We developed a general deep learning framework, FluidGAN, that is capable of learning and predicting time-dependent convective flow coupled with energy transport. FluidGAN is thoroughly data-driven with high speed and accuracy and satisfies the physics of fluid without any prior knowledge of underlying fluid and energy transport physics. FluidGAN also learns the coupling between velocity, pressure and temperature fields. Our framework could be used to learn deterministic multiphysics phenomena where the underlying physical model is complex or unknown.