Machine learning to predict plasma-based CO2 conversion in dielectric barrier discharges

Key Points

• The aim is to optimize plasma-based CO2 conversion processes using machine learning to better predict outcomes.
• Utilized machine learning algorithms to analyze data from dielectric barrier discharges.
• Tested various configurations to assess the prediction accuracy.
• Focused on reducing the complexity of the nonlinear behavior involved.
• Machine learning models achieved a prediction accuracy exceeding 85%, indicating strong potential.
• Significant improvement in efficiency compared to traditional experimental methods.
• Revealed key parameters affecting systemic behavior in plasma processes.

Abstract