< Back to previous page


Asymptotic-preserving Particle Methods for Kinetic Equations Arising in Plasma Edge Simulation

At present, numerical simulations of the plasma and neutral transport in the plasma edge and divertor region of nuclear fusion devices serve to interprete experimental observations and to design next generation reactors such as ITER and DEMO.
To this end, the B2-EIRENE code is used worldwide.
The plasma is modeled with fluid equations, while the neutral particles are described by a kinetic equation.
These two sets of equations are strongly coupled because of mutual interactions between plasma and neutral particles.
A Finite Volume (FV) method for the plasma equations and a Monte Carlo (MC) simulation of the neutral particles are iteratively coupled and solved alternately until convergence is attained.

Unfortunately, for large fusion devices the usability of the plasma edge simulations is limited because they require very long computation times and because convergence problems become apparent.
Nowadays these convergence issues, largely caused by the statistical noise of the MC method, are not well understood yet because of the complex coupling between the FV method and the MC method.
In this PhD project these coupled FV/MC iterative schemes are analyzed together with the corresponding convergence problems.
The first goal is to better understand the bias and variance on the solution.

Date:29 Aug 2016 →  15 Jul 2018
Keywords:Nuclear Fusion, Plasma Edge Simulation, Monte Carlo Methods
Disciplines:Applied mathematics in specific fields, Computer architecture and networks, Distributed computing, Information sciences, Information systems, Programming languages, Scientific computing, Theoretical computer science, Visual computing, Other information and computing sciences
Project type:PhD project