Stability-preserving quadrature-based moment-closure model hierarchy
报告人：黄骞（清华大学）
时间：2023-09-13 10:30-11:00
地点：理科一号楼1418
Abstract:
In this talk, I will present our recent efforts on the macroscopic moment-closure model hierarchy from kinetic equations. We focus on a quadrature-based approach which is positivity-preserving, of the conservative form and numerically efficient. However, the mathematical theory of the resultant moment-closure system is largely lacking, and it is difficult to extend the existing approaches to multidimensional velocities. To tackle the issues, we derive theoretical results on hyperbolicity, realizability and dissipative properties of the systems. Our results give clues to a theory-guided machine learning moment-closure approach. Then we develop new quadrature-based moment methods for several multidimensional scenarios, with applications to the rarefied gas flow and the polar active matter system.
个人简介：黄骞，清华大学能动系助研，于2017年毕业于清华大学获博士学位。主要从事动理学矩方法、清洁低碳燃烧技术等领域研究，累计发表核心期刊论文40余篇，包括SIAM J. Appl. Math.、J. Sci. Comput.、Combust. Flame等应用数学与燃烧能源领域期刊，授权发明专利7项，承担自然基金等十余项课题，曾获教育部自然科学一等奖（2022）、日内瓦国际发明展金奖（2023）等。
https://www.researchgate.net/profile/Qian-Huang-6
