“Physics-Driven Learning of the Steady Navier-Stokes Equations Using Deep Convolutional Neural Networks” (2022) Communications in Computational Physics, 32(3), pp. 715–736. doi:10.4208/cicp.OA-2021-0146.