Stark spectral line broadening modeling by machine learning algorithms

Various types of electric fields contained in the laboratory and astrophysical plasma cause a Stark broadening of spectral lines in plasma. Therefore, a large number of spectroscopic diagnostics of laboratory and astrophysical plasma are based on experimental and theoretical studies of Stark broaden...

Full description

Saved in:
Bibliographic Details
Published in:Neural computing & applications 2022-04, Vol.34 (8), p.6349-6358
Main Authors: Tapalaga, Irinel, Traparić, Ivan, Trklja Boca, Nora, Purić, Jagoš, Dojčinović, Ivan P.
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Science
Data analysis
Data Mining and Knowledge Discovery
Electric fields
Empirical analysis
Free electrons
Image Processing and Computer Vision
Laboratories
Line broadening
Line spectra
Machine learning
Original Article
Parameter identification
Plasma
Probability and Statistics in Computer Science
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Various types of electric fields contained in the laboratory and astrophysical plasma cause a Stark broadening of spectral lines in plasma. Therefore, a large number of spectroscopic diagnostics of laboratory and astrophysical plasma are based on experimental and theoretical studies of Stark broadening of spectral lines in plasma. The topic of the present investigation is the Stark broadening caused by free electrons in plasma and its dependence on certain atomic parameters using a new method based on the machine learning (ML) approach. Analysis of empirical data on atomic parameters was done by ML algorithms with more success that it was previously done by classical methods of data analysis. The correlation parameter obtained by artificial intelligence (AI) is slightly better than the one obtained by classical methods, but the scope of application is much wider. AI conclusions are applicable to any physical system while conclusions made by classical analysis are applicable only to a small portion of these systems. ML algorithms successfully identified quantum nature by analyzing atomic parameters. The biggest issue of classical analysis, which is infinite spectral line broadening for high ionization stages, was resolved by AI with a saturation tendency.
ISSN:0941-0643
1433-3058
DOI:10.1007/s00521-021-06763-4