Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas
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 gen...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Institute of Electrical and Electronics Engineers
2024
|
| Online Access: | https://hdl.handle.net/1721.1/155296 |
| _version_ | 1811094389570666496 |
|---|---|
| author | Datta, Rishabh Ahmed, Faez Hare, Jack D. |
| author_facet | Datta, Rishabh Ahmed, Faez Hare, Jack D. |
| author_sort | Datta, Rishabh |
| collection | MIT |
| description | 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. |
| first_indexed | 2024-09-23T15:59:16Z |
| format | Article |
| id | mit-1721.1/155296 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T15:59:16Z |
| publishDate | 2024 |
| publisher | Institute of Electrical and Electronics Engineers |
| record_format | dspace |
| spelling | mit-1721.1/1552962024-09-20T04:11:59Z Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas Datta, Rishabh Ahmed, Faez Hare, Jack D. 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. National Science Foundation (NSF) 2024-06-21T18:25:51Z 2024-06-21T18:25:51Z 2024 Article http://purl.org/eprint/type/JournalArticle 0093-3813 1939-9375 https://hdl.handle.net/1721.1/155296 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. 10.1109/TPS.2024.3364975 10.1109/tps.2024.3364975 Transactions on Plasma Science Creative Commons Attribution-Noncommercial-ShareAlike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers Author |
| spellingShingle | Datta, Rishabh Ahmed, Faez Hare, Jack D. Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas |
| title | Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas |
| title_full | Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas |
| title_fullStr | Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas |
| title_full_unstemmed | Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas |
| title_short | Machine-Learning-Assisted Analysis of Visible Spectroscopy in Pulsed-Power-Driven Plasmas |
| title_sort | machine learning assisted analysis of visible spectroscopy in pulsed power driven plasmas |
| url | https://hdl.handle.net/1721.1/155296 |
| work_keys_str_mv | AT dattarishabh machinelearningassistedanalysisofvisiblespectroscopyinpulsedpowerdrivenplasmas AT ahmedfaez machinelearningassistedanalysisofvisiblespectroscopyinpulsedpowerdrivenplasmas AT harejackd machinelearningassistedanalysisofvisiblespectroscopyinpulsedpowerdrivenplasmas |