Seminars

                                                      OpenFoam Day

                                                          Pictures of the Day

On Januari 10, 2012 there will be two events in which you can participate:

1. The validation workshop of the SIGs on Martitime
   Applications/FSI and Multiphase
2. The Dutch OpenFOAM user meeting

Validation workshop

One of the main issues confronting OpenFOAM user is the lack of available validation results. Therefore, two Special Interest Groups (Maritime applications/FSI and Multiphase) of the Dutch OpenFOAM users group have decided to start a validation initiative.

The initiative will aim at:

• determining how reliable OpenFOAM is for real-life test cases
• learning how to use OpenFOAM in order to obtain best results
• identifying issues for further research

The meeting is intended for people who actually want to contribute to the validation activities (in the framework of their research or because they need it for their work). Lunch is available from 13:00 to 14:00 for those who also want to participate in the general meeting of the Dutch OpenFOAM user meeting in the afternoon.

User meeting

In the afternoon the keynote speaker for this event is Mattijs Janssens from OpenCFD, one of the two main companies that are developing OpenFOAM.
The meeting will take place right after the validation workshop, (please see below the program).

Participation in both events is free. Please register before January 3 2012 on http://www.aanmelder.nl/openfoam3

Program

Fluid Flow Simulations: from Particles to PDE's

Thursday, November 10, 2011

Delft University of Technology
Faculty of EEMCS, Lecture room D, Mekelweg 4, Delft 

Organizer: Delft Centre for Computational Science and Engineering

Registration: Please send an e-mail to Deborah Dongor (D.M.Dongor@remove-this.tudelft.nl)
                     There is no registration fee

Program

13:45 Coffee

14:00 Opening by Prof.dr.ir. Chris Kleijn

14:05 Prof. Tom Schwartzentruber (Department of Aerospace Engineering and Mechanics, University of Minnesota, USA) 

14:35 Discussion

14:40 Ira Livshits (Department of Mathematical Sciences, Ball State University, USA) 

15:10 Discussion

15:15 Coffee break

15:30 Dr. Jurriaan J. J. Gillissen (Department of Mathematical Sciences, Ball State University, USA) 

16:00 Discussion

16:05 Closing by Prof.dr.ir. Chris Kleijn

Speaker 1. Prof. Tom Schwartzentruber,
Department of Aerospace Engineering and Mechanics, University of Minnesota, USA

Particle Simulation of Nonequilibrium Hypersonic Flows
During hypersonic reentry, a high temperature gas in thermal and chemical nonequilibrium surrounds the vehicle. Understanding the precise thermo-chemical state within the shock layer, boundary layer, and at the vehicle surface is a necessary first step in predicting processes such as radiative heating and catalytic or ablative processes occurring on the heat shield surface. In addition to chemical and thermal nonequilibrium, the gas may also be in a state of collisional-nonequilibrium where gas molecules may not have equilibrium (Maxwell-Boltzmann) velocity and internal energy distributions. An accurate approach for modeling such flows is the direct simulation Monte Carlo (DSMC) particle method. The DSMC method moves a representative number of simulated molecules through a computational mesh, allowing for collisions including those with the vehicle surface. If the correct collision rates, probabilities of internal energy exchange and chemical reactions, as well as collision outcomes are prescribed, the DSMC method can accurately simulate complex hypersonic flows.

Specifically, DSMC collision models are typically parameterized by forcing consistency with existing continuum models in the limit of continuum flow (Maxwell-Boltzmann molecular distributions). Examples including models for viscosity, thermal conductivity, vibrational relaxation, and finite rate chemistry are highlighted in context with hybrid CFD-DSMC simulations. In some cases, when the continuum models must be interpreted from relatively few experimental data sets and are often extrapolated far outside of experimental conditions, their use for informing DSMC collision models is questionable.

Instead, Computational Chemistry simulations are employed that require only a Potential Energy Surface (PES) describing each collision and do not rely on collision rules or constitutive closure laws. A novel combined event-driven/time-driven numerical technique will be described that greatly accelerates the first-principles simulation of dilute gas flows. Simulation results are presented for normal shock waves and are compared with experimental data and DSMC simulations with regards to viscosity, mass diffusion, and rotational excitation processes within the shock.

Speaker 2. Dr. Ira Livshits
Department of Mathematical Sciences, Ball State University, USA

Application of algebraic multigrid approach to solving the Helmholtz equation with large wave numbers
Multigrid algorithms are known to be efficient for many  partial differential equations. However, Helmholtz equations with large wave numbers posses properties that standard iterative solvers do not handle well. In this talk, we discuss the challenges that the Helmholtz equations present and some ways to overcome them using multigrid, including algebraic multigrid, techniques.

Speaker 3. Dr. Jurriaan J. J. Gillissen
Department of Multi-Scale Physics, Delft University of Technology, The Netherlands

Lattice Boltmann Simulations of Turbulent Drag Reduction using Spherical Additives
We have used lattice Boltzmann and immersed boundary methods to simulate turbulent Couette flow of suspensions of particles whose radii are large compared to the viscous sub-layer VSL thickness. Drag reduction is observed. We explain the drag reduction by considering the finite size of the particles. Due to their large size, the particle volume fraction decays to zero in the VSL, when approaching the wall. Consequently, the particles induce a negligible effect in the VSL, while at the same time they dampen the eddies in the turbulent region. This mechanism is similar to the case of polymers, which exemplifies the universal nature of drag reduction.