Physics-informed neural networks (PINN) – neural networks that are trained to solve supervised learning tasks while respecting any given laws of physics described by general nonlinear partial differential equations – are quickly developed in recent years since they are introduced by Raissi and Karniadakis in the context of solving partial differential equations. We decided to introduce PINN into fluid mechanics, where the basic equations, Navier–Stokes equations, are partial differential equations. Further development is still in progress.

Division of Mechanics, Beijing Computational Science Research Center