Predicting Plasma Temperature From Line Intensities Using ML Models

TL;DR

This study demonstrates that various machine learning models, especially Random Forest, can accurately predict plasma temperature from spectral line intensities, capturing complex relationships in plasma diagnostics.

Contribution

The paper compares multiple ML models for plasma temperature prediction and finds Random Forest performs best, highlighting the effectiveness of ML in plasma diagnostics.

Findings

Random Forest achieved the highest accuracy among models.

ML models effectively capture complex line-intensity and temperature relations.

High prediction accuracy demonstrated for plasma temperature using ML.

Abstract

In this work, ML models were used to predict the plasma temperature using the dataset obtained by implementing the CR-model for Na-like Krypton. The models included in the study are: Linear Regression, Lasso Regression, Support Vector Regression, Decision Trees, Random Forest, XGBoost, Multi-layer Perceptron and Convolutional Neural Network. For evaluating the models we used Mean Absolute Error, Mean Squared Error and R^2 Score as metrics, In our study Random Forest performed best as compared to other model considered, the study conclude that complex relation between the line-intensities and Plasma temperature can be capture by ML models and they can be used to predict the temperature with high accuracy.