Machine learning in the study of phase transition of two-dimensional complex plasmas

TL;DR

This paper demonstrates how machine learning, specifically convolutional neural networks, can effectively identify phase transitions in two-dimensional complex plasmas by analyzing particle images from simulations and experiments.

Contribution

It introduces a novel approach combining Langevin dynamics simulations with CNN-based image analysis to map phase diagrams and study melting in complex plasmas.

Findings

CNN accurately identifies phase states from particle images

The method produces reliable phase diagrams

Applicable to experimental video microscopy data

Abstract

Machine learning is applied to investigate the phase transition of two-dimensional complex plasmas. The Langevin dynamics simulation is employed to prepare particle suspensions in various thermodynamic states. Based on the resulted particle positions in two extreme conditions, bitmap images are synthesized and imported to a convolutional neural network (ConvNet) as training sample. As a result, a phase diagram is obtained. This trained ConvNet model can be directly applied to the sequence of the recorded images using video microscopy in the experiments to study the melting.