セミナーシリーズ概要

シミュレーション科学の力による流体現象の解明を行うComputational Fluid Dynamics (CFD)とともに、データ科学の力を借りる Data-driven Fluid Dynamics (DFD) の進展も目覚ましい。 東京工業大学大西研究室では、CFDだけでなく、DFDによる環境流体研究の新たな展開を目指している。 このセミナーシリーズでは、関連分野の最新研究をおこなっている研究者に話題を提供していただく。

発表者: Thibault Moli MAUREL OUJIA,
1)Ph.D student, Institut de Mathématiques de Marseille (I2M), Aix-Marseille University, CNRS, France
2)External researcher, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Japan

開催日時: 2023年8月2日(水)16:00-17:00
開催場所: 東工大大岡山キャンパス石川台9号館208室 (アクセス)
タイトル

A tessellation-based approach to study the dynamics of inertial particles in turbulence

アブストラクト

Inertial particle laden turbulent flows occur in many natural and engineering systems, as droplets in clouds in the atmosphere, sandstorms in the desert or spray combustion with liquid fuels. Understanding of the complex dynamics of these phenomena is a prerequisite for sound modeling and important for numerous applications. We investigate the behavior and dynamics of heavy particles in particle-laden homogeneous isotropic turbulence using tessellation techniques. We introduce a tessellation-based method that allows us to study the dynamics of particles. This method, which uses a modified Voronoi tessellation, assesses volume dynamics in three-dimensional moving particle clouds. The temporal rate of change of the volumes yields the divergence of the particle velocity and by rearranging the coefficients of the velocity, we can compute the curl and the velocity gradient tensor. Applying this novel approach to the study of both fluid and inertial particles in fully developed three-dimensional isotropic turbulence, we explore the divergence, convergence, and vortical behavior of particle motion. This approach, demonstrating first-order convergence in both space and time, enhances our understanding of inertial particle clustering in complex turbulence phenomena. Furthermore, a novel multiscale approach for analyzing inertial particle dynamics is presented, focusing on the divergence, by leveraging the tessellation-based technique. This multiscale analysis enables a deeper comprehension of the formation and dynamics of particle clusters.