Lecture: Different Perspectives on the Application of Deep Learning for Pore-Scale Two-Phase Flow

Lecturer: Prof. Jiang Zeyun (Heriot-Watt University)

Time: 16:30-18:30, June 8^th.

Venue: C 302B, Minglilou Building

Abstract:

The recent pore-scale literature is not short of compelling studies on deep learning applications. The prediction of two-phase flow fields, however, remains elusive. This is partly due to focusing on model architecture and data quality, rather than the quantity of data. This work presents an end-to-end and accurate deep learning workflow to predict phase distributions during steady-state two-phase drainage, directly from dry images and input features of pixel size, IFT, contact angle, and pressure. A highly diverse dataset is first constructed by subsampling CT scans of synthetic and real rocks. We then devise a new vision transformer (ViT) that drains pores solely based on their size, regardless of their spatial location, where the phase connectivity to inlet(s) is enforced as a post-processing step. With this setup, inference on images of any size with various pixel sizes can efficiently be made by patching input images and stitching results.

Bio:

Prof. Jiang Zeyun has been mainly engaged in the analysis of heterogeneous multi-scale structures of porous media (such as rocks, soils, etc.) and the study of fluid seepage models since 2004. He has published more than 40 papers in academic journals such asWater Resource research,Transport in Porous Media,Fueland has attended some major international conferences. Professor Jiang Zeyun has led and participated in several major scientific research projects both at home and abroad.

Organizer and sponsor:

School of Sciences

Institute of Artificial Intelligence

Institute of Nonlinear Dynamical Systems

Mathematical Mechanics Research Center

Institute of Science and Technology Development