数学科学系

Department of Mathematical Sciences

Accelerating the self-consistent field iteration for solving the Kohn-Sham equation

报告人:高兴誉(副研究员,北京应用物理与计算数学研究所、中物院高性能数值模拟软件中心)

时间:2018927日周四下午4:00-5:00

地点:理科楼数学系A-404

联系人:黄忠亿、吴昊

摘要:Solving the Kohn-Sham equation is most computationally demanding in the first-principles calculations of materials. The self-consistent field (SCF) iteration is used to solve such a nonlinear eigenvalue problem. We develop techniques to accelerate the convergence of SCF iteration in static calculations and Born-Oppenheimer molecular dynamics (BOMD) separately. In the static calculation or the first step of BOMD, we modify the Kerker preconditioning scheme to capture the long-range screening behavior of inhomogeneous systems. The effectiveness and efficiency is shown by the tests on long-z slabs of metals, insulators, and metal-insulator contacts. In the following BOMD simulation, the wavefunction extrapolation greatly reduces the number of SCF iterations. Going against the intuition that the higher order of extrapolation possesses a better accuracy, we demonstrate that there exists an optimal extrapolation order in terms of minimal number of SCF iterations. It is illustrated that the optimal extrapolation order covers a broad range when varying the MD time step or the SCF convergence criterion.

报告人简介:2009年毕业于中科院数学与系统科学研究院,获博士学位。现就职于北京应用物理与计算数学研究所高性能计算中心、中物院高性能数值模拟软件中心金属材料团队,主要从事大规模第一性原理模拟的计算方法研究。2011年和2013年两获全国高性能计算学术年会优秀论文奖,2015年获军队科技进步一等奖。