报告人：Prof. Chunlei Liang (梁春雷) The George Washington University
报告人简介：Dr. Chunlei Liang is an Associate Professor of Engineering and Applied Science at The George Washington University. He is an editorial board member of Computers & Fluids, an Elsevier Journal and an Associate Fellow of the American Institute of Aeronautics and Astronautics (AIAA). Prof. Liang received a Young Investigator Program award from the Office of Naval Research in 2014 and a CAREER award from the US National Science Foundation in 2016.
报告一: High-order Spectral Difference Method for Studying Rotary Wing Aerodynamics and Thermal Convection and Magneto-hydrodynamics for the Sun
摘要：Two recent advancements of high-order spectral difference (SD) method for computational fluid dynamics on unstructured meshes will be presented. The first progress is our contribution to a new curved sliding-mesh approach to the SD method for simulating rotary wing aerodynamics. The second elevation of the SD method is our recent successful design of a massively parallel code, namely CHORUS, for predicting thermal convection in the Sun. Recently, we have also built a simulation capability for predicting magneto-hydrodynamics of the Sun.
报告二：A High-order Sliding and Deforming Spectral Difference Method for Exascale Simulations of Turbulent Flows
摘要:I will present a novel high-order sliding and deforming spectral difference (SD^2) method for solving compressible Navier-Stokes equations on unstructured quadrilateral grids. The SD^2 method is an extension of the sliding-mesh spectral difference method (Zhang & Liang, 2015, J. Computational Physics) to coupled rotating and deforming domains. Through a simple sliding-mesh interface, the SD^2 method mitigates large grid distortion which is often resulted from rotating wall boundaries. Meanwhile, the SD^2 algorithm adopts an arbitrary Lagrangian-Eulerian framework that can handle large-amplitude translational motions of grid points on deforming domains. This new SD^2 solver is verified by using several benchmark flow problems able to demonstrate optimal orders of accuracy in space. The SD^2 method is efficient and robust for all our test problems and is expected to be a competitive algorithm for simulating turbulent flows of wind turbines by using exascale computers in the near future.