Bremen 2017 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

P: Fachverband Plasmaphysik

P 3: Dusty Plasmas I

P 3.4: Talk

Monday, March 13, 2017, 15:10–15:25, HS 1010

A Machine-Learning Method for Plasma Crystal Analysis — •Christopher Dietz, Tobias Kretz, and Markus Thoma — I. Physikalisches Institut, JLU Gießen

Machine-Learning is one of the most popular fields in computer science and has a vast amount of applications. We propose a method, that will use a Machine-Learning algorithm to identify crystal structures in a simulated plasma crystal. This data consists of fcc, hcp, bcc and disordered particles, which is difficult to identify using conventional methods.

The new approach works very well for highly disturbed lattices and mixed phase systems. Thus, it enables us to retrieve a more complete picture of plasma crystals.