Lagrangian and multi-scale investigations of turbulent flows

时间:2013-10-28浏览:32

报告题目:Lagrangian and multi-scale investigations of turbulent flows
报告人:杨  越 研究员
单位:北京大学工学院
报告时间:2013年9月25日(周三)14:30
报告地点:工程学院力一楼239会议室

报告人简介:
杨越,2004年浙江大学获学士学位,2007年中国科学院力学所获硕士学位(导师:何国威研究员),2011年美国加州理工学院获博士学位(导师:Dale Pullin教授)。其博士研究工作为流体湍流与涡动力学发展了新的拉格朗日理论框架与多尺度诊断工具。博士后期间获美国普林斯顿大学燃烧能源中心资助,于康奈尔大学和桑迪亚国家实验室开展湍流燃烧数值模拟方面的研究工作(导师:Stephen Pope教授)。2013年加入北京大学工学院,现任特聘研究员、博士生导师。

报告摘要:
The Lagrangian method for the study of turbulence is a classical but challenging approach. We present a series of Lagrangian investigations of  turbulent flows with increasing complexities, from the dispersion of tracer particles to the evolution of Lagrangian material surfaces, from Lagrangian kinematics to Lagrangian vortex dynamics, and from the subgrid-scale modeling of inert turbulent mixing to the large-eddy simulation (LES) of  turbulent combustion with Lagrangian particle methods. The goal is to develop novel computational and multi-scale methods for quantitative  predictions of turbulent mixing and combustion. In particular, we study the non-local geometry of finite-sized Lagrangian structures in homogenous isotropic turbulence and wall-bounded turbulence. The multi-scale geometric analysis is applied on the evolution of Lagrangian fields, to extract Lagrangian structures at different length scales and to characterize their non-local geometry in a space of reduced geometrical parameters. In order to explore the connection and corresponding representations between Lagrangian kinematics and vortex dynamics, we develop a theoretical formulation and  numerical methods for computation of the evolution of a vortex-surface field. In the Lagrangian framework, we apply the probability density function (PDF)  approach to turbulent combustion with a LES/PDF modeling study of a non-premixed CO/H2 temporally-evolving turbulent planar jet flame. The performance of the hybrid LES/PDF methodology is assessed through detailed a posteriori comparisons with direct numerical simulation of the same flame.