Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas
Name
Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas.pdf
Size
2.76 MB
Format
Adobe PDF
Checksum (MD5)
e9ba6bfaf104038fba19774c67d3f350
Author(s)
Datta, Rishabh
Ahmed, Faez
Hare, Jack D.
Date Issued
2024
Journal
Transactions on Plasma Science
Publisher
Institute of Electrical and Electronics Engineers
Citation
R. Datta, F. Ahmed and J. D. Hare, "Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas," in IEEE Transactions on Plasma Science.
Version
Author's final manuscript
Abstract
We use machine-learning (ML) models to predict ion density and electron temperature from visible emission spectra, in a high-energy density pulsed-power-driven aluminum plasma, generated by an exploding wire array. Radiation transport simulations, which use spectral emissivity and opacity values generated using the collisional-radiative code PrismSPECT, are used to determine the spectral intensity generated by the plasma along the spectrometer’s line of sight (LOS). The spectra exhibit Al-II and Al-III lines, whose line ratios and line widths vary with the density and temperature of the plasma. These calculations provide a 2500-size synthetic dataset of 400-D intensity spectra, which is used to train and compare the performance of multiple ML models on a three-variable regression task. The AutoGluon model performs best, with an R2 -score of roughly 98% for density and temperature predictions. Simpler models random forest (RF), k -nearest neighbor (KNN), and deep neural network (DNN) also exhibit high R2 -scores ( > 90% ) for density and temperature predictions. These results demonstrate the potential of ML in providing rapid or real-time analysis of emission spectroscopy data in pulsed-power-driven plasmas.
MIT Department
Massachusetts Institute of Technology. Plasma Science and Fusion Center
Massachusetts Institute of Technology. Department of Mechanical Engineering
Terms of Use
Creative Commons Attribution-Noncommercial-ShareAlike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1109/tps.2024.3364975
