Homepage

Jannis Teunissen

research:start

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revisionBoth sides next revision
research:start [2019/04/08 19:48] – [List of research ideas] jannisresearch:start [2019/12/03 16:22] – external edit 127.0.0.1
Line 1: Line 1:
 ===== Research ===== ===== Research =====
  
-==== Computational physics ====+==== Experience and interests ==== 
 + 
 +My general research interests are scientific computing, computational (plasma) physics and more recently also machine learning / data science.
  
 I have developed both particle-in-cell and plasma fluid codes for the simulation I have developed both particle-in-cell and plasma fluid codes for the simulation
-of electric discharges (non-thermal plasmas), see [[projects:start]]. I'm also +of electric discharges (non-thermal plasmas). I'm also active in the development 
-active in the development of [[http://amrvac.org/|MPI-AMRVAC]], a framework for +of [[http://amrvac.org/|MPI-AMRVAC]], a framework for (magneto)hydrodynamics 
-(magneto)hydrodynamics simulations. From a computational point of view, my work +simulations. My work has focused on topics such as adaptive mesh refinement 
-has focused on topics such as:+(AMR) and fast elliptic solvers (e.g., multigrid). I like to study systems that 
 +have some intrinsic complexity, not coming from boundary conditions or input 
 +data.
  
-  * Adaptive mesh refinement (AMR) +Since 2018, I have been working on machine learning methods applied to space weather applications, in collaboration with [[https://ecamporeale.github.io/|Enrico Camporeale]], as I have taken over two EU projects in this direction from Enrico (AIDA and ESCAPE, see below). Current research focuses on forecasting time-series data, recognizing magnetic reconnection, and the use of unsupervised methods for e.g. dimensionality reduction and clustering.
-  * Parallelization (using MPI and/or OpenMP) +
-  * Fast elliptic solvers (multigrid) +
-  * Adaptive control of particle weights+
  
-==== Machine learning ====+=== Research projects ===
  
-Since the end of 2018 I'm involved in two projects on machine learning for space +  * [[https://www.cwi.nl/news/2019/technology-grant-for-plasma-for-plants-project|Plasma for Plants (TTW)]] 
-physics applications, namely [[http://www.aida-space.eu/|AIDA]] and +  * [[https://www.nwo.nl/en/research-and-results/research-projects/i/64/27064.html|Let CO2 spark! (TTW)]] 
-[[https://cordis.europa.eu/project/rcn/219246/factsheet/en|ESCAPE]], together +  * [[https://www.cwi.nl/news/2018/technology-grant-awarded-to-optimize-plasma-assisted-combustion|Plasma assisted combustion (TTW)]] 
-with [[https://homepages.cwi.nl/~camporea/|Enrico Camporeale]].+  * [[http://www.aida-space.eu/|AIDA (H2020)]] 
 +  * [[https://cordis.europa.eu/project/rcn/219246/factsheet/en|ESCAPE (H2020)]]
  
-==== List of research ideas ====+==== Simulation codes ====
  
-This is an incomplete list of topics that I'd like to investigate. If you are +These are some of the simulation codes that I have developed or worked on:
-interested in collaborating or doing a student project, feel free to get in +
-touch (see [[contact:start]]).+
  
 +  * [[http://amrvac.org|MPI-AMRVAC]]
 +  * [[https://gitlab.com/MD-CWI-NL/afivo|Afivo]]
 +  * [[https://gitlab.com/MD-CWI-NL/afivo-streamer|Afivo-streamer]]
 +  * [[https://github.com/jannisteunissen/octree-mg|Octree-mg]]
 +  * [[https://gitlab.com/MD-CWI-NL/particle_swarm|Particle_swarm]]
 +  * [[http://cwimd.nl/doku.php?id=codes:particle|pamdi3d]]
 +  * [[https://github.com/jannisteunissen/streamer_1d|streamer_1d]]
 +
 +And these are some of the (simulation) utilities that I have developed:
 +
 +  * [[https://github.com/jannisteunissen/config_fortran|config_fortran]]
 +  * [[https://github.com/jannisteunissen/lookup_table_fortran|lookup_table_fortran]]
 +  * [[https://github.com/jannisteunissen/rng_fortran|rng_fortran]]
 +  * [[https://github.com/jannisteunissen/morton_fortran|morton_fortran]]
 +
 +==== Ideas for (small) projects ====
 +
 +A (very!) incomplete list of research ideas, some of which are suitable for student projects:
 +
 +  * Solving plasma fluid equations implicitly. In particular, what is a good preconditioner?
 +  * Improving the convergence rate of Monte Carlo particle swarm simulations in low electric fields
 +  * Coupling explicit and implicit time integration for plasma fluid models
   * Adding support for internal boundary conditions in a geometric multigrid solver.   * Adding support for internal boundary conditions in a geometric multigrid solver.
-  * Finding the fastest possible direct sparse method for solving the coarse grid equations in a geometric multigrid method.+  * Exploring efficient methods for solving the coarse grid equations in a geometric multigrid method.
   * Enabling efficient visualization of octree AMR data in Visit or Paraview.   * Enabling efficient visualization of octree AMR data in Visit or Paraview.
   * Coupling stiff chemistry to simulations with AMR (adaptive mesh refinement), where the chemistry can be evaluated at coarser resolution and perhaps partially implicitly.   * Coupling stiff chemistry to simulations with AMR (adaptive mesh refinement), where the chemistry can be evaluated at coarser resolution and perhaps partially implicitly.
   * Performing large scale 3D simulations of [[https://en.wikipedia.org/wiki/Sprite_(lightning)|sprite]] formation   * Performing large scale 3D simulations of [[https://en.wikipedia.org/wiki/Sprite_(lightning)|sprite]] formation
   * Coupling particle and fluid models in energy space, for the study of [[https://en.wikipedia.org/wiki/Runaway_electrons|runaway electron]] production in electric discharges.   * Coupling particle and fluid models in energy space, for the study of [[https://en.wikipedia.org/wiki/Runaway_electrons|runaway electron]] production in electric discharges.
-  * More robustly solving plasma fluid equations in discharge models using explicit time integration. 
   * Investigating the so-called "stability field" of streamer discharges through computations, with the goal of predicting how this field depends on the gas.   * Investigating the so-called "stability field" of streamer discharges through computations, with the goal of predicting how this field depends on the gas.
-  * Investigating what happens when discharges are 'long' compared to the time scale for electron attachment. +  * Extending the discharge model comparison of this [[http://dx.doi.org/10.1088/0963-0252/24/6/065002|paper]] to other fluid models
-  * Making a tool to estimate discharge inception probability and jitter from an electrostatic potential (//currently in progress with Shahriar Mirpour and Andy Martinez//).+
   * Compare particle-in-cell and plasma fluid models for 2D and 3D simulations of streamer discharges.   * Compare particle-in-cell and plasma fluid models for 2D and 3D simulations of streamer discharges.
- +  * Using DSMC (or similar particle-based simulations) to evaluate (and improve?) the behavior of continuum hydrodynamics schemes at low pressures/densities 
-==== Publications ==== +  * Exploring methods to make conventional hydrodynamics schemes more robust (i.e., avoiding negative pressures)
- +
-[[publications:start|List of publications]]. +
- +
-==== PhD ==== +
- +
-I finished my [[publications:thesis|PhD thesis]] in November 2015. The work was done in +
-the [[http://cwimd.nl|Multiscale Dynamics]]((This is our official (CWI-hosted) +
-[[https://www.cwi.nl/research-groups/Multiscale-Dynamics|webpage]])) group at +
-CWI, under the supervision of [[http://homepages.cwi.nl/~ebert/|Ute Ebert]]. +
- +
-==== General interests ==== +
- +
-After finishing secondary school in 2005, there were three topics that I wanted +
-to study: +
- +
-   * How a (human) brain works +
-   * How a computer works +
-   * How nature works +
- +
-Thus far, I have mostly been busy with the latter two questions, with my primary +
-research interests being computational science and computational physics. I like +
-to study systems that have some intrinsic complexity, not coming from boundary +
-conditions or input data. +
  

Page Tools