应能源与动力工程学院刘海湖教授的邀请，英国思克莱德大学的苏微博士将于2017年10月24日至26日来我校进行学术访问，并针对直接求解Boltzmann模型方程的高阶有限元方法进展进行一次公开学术讲座，届时欢迎各位师生参与交流。

讲座题目：直接求解Boltzmann模型方程的高阶有限元方法

讲座时间：2017年10月25日 上午10:00 -12:00
讲座地点：北二楼1401会议室
讲座人：苏微 博士

讲座内容摘要：

The nonequilibrium rarefied gas flows are often encountered in the high-altitude aerodynamics, the nano/micro-electro-mechanical systems (N/MEMS), and vacuum technology. Accurate physical models and efficient numerical methods are required for predicting the nonequilibrium flow phenomena encountered in rarefied gas flow conditions. The mathematical theory of the rarefied gas dynamic is the kinetic theory, in which the evolution of the gas molecules is described by the Boltzmann equation. The direct simulation Monte Carlo (DSMC) method is the most widely used numerical method to solve the Boltzmann equation. However, the major features of this stochastic method make it inefficient in simulating low-speed flows and unsteady flows, and hinder coupling to continuum CFD and deterministic structural, thermal and electrostatic modeling. The deterministic numerical simulation (DNS) approaches relaying on the discrete ordinate method (DOM) are attractive alternatives to overcome the DSMC limitations. These methods adopt a numerical quadrature to approximate the integration with respect to molecular velocity on a discrete set of velocities. Then, the distribution functions which are continuous in physical space and time but discrete in velocity space are solved by various of CFD methods.

   This talk discusses the development of a high-order Runge-Kutta Discontinuous Galerkin (RKDG) method for the deterministic solution of the Boltzmann kinetic equation. A conservative DG type discretization of the nonlinear collision relaxation term is formulated for both the Bhatnager-Gross-Krook and the ellipsoidal statistical kinetic models. The RKDG method has up to 3^rd order spatial accuracy and up to 4^th-order time accuracy. The computational performance of the RKDG method is compared with a widely used 2^nd order finite volume method (FVM). It is shown that 2^nd order RKDG is over 15 times faster than the 2^nd order FVM method for the Couette flow test case.  

讲座人简介：
苏微博士现在是英国思克莱德大学James Wier流体实验室的Visiting Researcher。她于2015年获得北京航空航天大学的工学博士学位。她的研究兴趣包括计算稀薄气体动力学方法及其应用，高马赫数激波后气体振动非平衡-化学反应-电离耦合。目前为止，苏微博士已经在Physics Review E, Physics of Fluids等国际期刊发表SCI论文7篇。