Kai Zhang

TL;DR: A deep-learning-based model is developed to solve three categories of problems controlled by the subsurface 2D oil/water two-phase flow PDE based on the FNO, a recently proposed high-efficiency PDE solution architecture that overcomes the shortcomings of popular algorithms such as physics-informed neural networks and fully convolutional network.

TL;DR: The results show that the well pattern can perfectly adapt to anisotropic and uncertain reservoirs, and the optimization method can remarkably reduce the computational amount while ensuring the accuracy of the results.

TL;DR: The results indicate that the well-placement optimization proposed in this paper can improve the performance of an EGS.

TL;DR: In this paper, a coupled hydro-mechanical model for simulating the complex behavior of fractured and karstified porous media is developed, where a two-phase Darcy's equation is used to describe fluid flow in both matrix and fractures, while the free flow in cavities is considered based on an assumption of phase instantaneous gravity segregation.

TL;DR: In this article, the authors proposed a new method for assessing the risk of collapse in subway tunnels based on the Delphi method, which is applied to an actual project to verify its correctness, and the prediction results have good consistency with the actual tunnel.