报告题目：The Efficient Conservative S-DDM Method for Environmental Modelling
报告时间：5月14日（周二）下午3:30—5:00
报告地点：尚贤楼 706报告厅
主持人：王廷春副教授
报告人简介：梁栋，加拿大约克大学终身正教授，计算数学、环境和能源数值模拟专家，加拿大CMS、CAIMS和国际SIAM学会会员，山东省泰山学者海外特聘教授。梁栋教授在数值分析与科学计算、多孔介质渗流力学计算方法、大气环境计算、地下水污染计算和控制、海水入侵、计算电磁学和计算生物等研究领域取得了许多原创性成果，在《J. Comput. Phys.》、《Numer. Math.》、《SIAM J. Numer. Anal.》、《SIAM J. Sci. Comput.》、 《IMA J. Numer. Anal.》、《IMA J. Appl. Math.》、《J.Aero. Sci.》、《Atmos. Environ.》等国际权威学术杂志上发表论文120余篇，先后承担加拿大国家自然科学基金（NSERC）项目、MITACS项目和中国“973”大气环境项目等多项研究课题。2012年获加拿大NSERC的DAS奖励。欢迎各位老师和同学参与交流！
报告摘要：Environment computation is an important and difficult problem. Mathematical models describing multi-component pollution flows in porous media and for atmospheric environment are the time-dependent and coupled nonlinear partial differential equations. Due to the complexity, the nonlinearity, and the large scale and long term of simulation and protection, it is important to develop efficient numerical methods for solving multicomponent pollution flows in parallel computing. In this talk, I will present our conservative S-DDM method that we develop for computing nonlinear multicomponent contamination flows over multiple non-overlapping domain decomposition. The significant feature of our scheme is that while it keeps the excellent advantages of the non-overlapping domain decomposition and the splitting technique, the developed algorithm conserves mass over the whole domain. Numerical experiments show the excellent performance of our proposed conservative method for computing nonlinear multicomponent pollution flows in environmental computation including atmospheric pollution.
数学与统计学院
2019年5月10日
