Predicting Discharge Mode in Dielectric Barrier Discharge (DBD) Systems using Machine Learning: Importance, Algorithms, and Perspectives

Abstract

This paper discusses the importance of predicting discharge modes in dielectric barrier discharge (DBD) plasma systems, which are critical for optimizing their performance and ensuring their safe operation. Several parameters can influence the characteristics of the discharge, including the applied voltage, gas pressure, gas temperature, electrode configuration, dielectric material, and gas type. The discharge mode, which can be filamentary or homogenous, is one of the crucial elements impacting the performance of DBD devices. The paper highlights the potential of machine learning for predicting the discharge mode in DBD systems and discusses the main challenges related to data collection in this area. The paper aims to explore ways to collect more data to train more accurate and reliable machine learning models for predicting discharge modes in DBD systems.