Image-Based Methods to Investigate Synchronization between Time Series Relevant for Plasma Fusion Diagnostics

Area of Science:

• Fusion energy research
• Plasma physics
• Time series analysis

Background:

• Advanced time series analysis and causality detection are crucial for assessing synchronization experiments in tokamaks.
• Lag synchronization is a key strategy for controlling fusion plasma instabilities using pace-making techniques.
• Evaluating pace-making efficiency is challenging due to causal effects coexisting with plasma instability periodicity.

Purpose of the Study:

• To investigate image representation methods for evaluating the efficiency of pace-making techniques in fusion plasma control.
• To introduce and assess novel image-based approaches for analyzing time series data from tokamak experiments.

Main Methods:

• Utilized Gramian Angular Field (GAF), Markov Transition Field (MTF), and Chaos Game Representation (CGR) for time series image representation.
• Proposed an original variation of the Markov Transition Matrix for analyzing coupled time series.
• Incorporated a cross-visibility network mapping method to represent time series as images.

Main Results:

• Evaluated the performance of GAF, MTF, CGR, and the novel Markov Transition Matrix variation on synthetic data.
• Applied the developed methods to analyze real-world data from JET tokamak experiments.
• Demonstrated the potential of image-based time series analysis for assessing fusion plasma control strategies.

Conclusions:

• Image representation techniques offer a promising avenue for evaluating the effectiveness of pace-making in fusion plasma control.
• The proposed methods, including the novel Markov Transition Matrix, provide valuable tools for analyzing complex plasma dynamics.
• Successful application to JET data highlights the practical utility of these advanced analytical approaches in fusion research.