Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques

This article presents a new algorithm that uses techniques from the field of artificial intelligence to automatically estimate the physical parameters of massive stars from a grid of stellar spectral models. This is the first grid to consider hydrodynamic solutions for stellar winds and radiative tr...

Full description

Saved in:
Bibliographic Details
Main Authors: Esteban Olivares, Michel Curé, Ignacio Araya, Ernesto Fabregas, Catalina Arcos, Natalia Machuca, Gonzalo Farias
Format: Article
Language:English
Published: MDPI AG 2024-10-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/12/20/3169
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850206093265862656
author Esteban Olivares
Michel Curé
Ignacio Araya
Ernesto Fabregas
Catalina Arcos
Natalia Machuca
Gonzalo Farias
author_facet Esteban Olivares
Michel Curé
Ignacio Araya
Ernesto Fabregas
Catalina Arcos
Natalia Machuca
Gonzalo Farias
author_sort Esteban Olivares
collection DOAJ
description This article presents a new algorithm that uses techniques from the field of artificial intelligence to automatically estimate the physical parameters of massive stars from a grid of stellar spectral models. This is the first grid to consider hydrodynamic solutions for stellar winds and radiative transport, containing more than 573 thousand synthetic spectra. The methodology involves grouping spectral models using deep learning and clustering techniques. The goal is to delineate the search regions and differentiate the “species” of spectra based on the shapes of the spectral line profiles. Synthetic spectra close to an observed stellar spectrum are selected using deep learning and unsupervised clustering algorithms. As a result, for each spectrum, we found the effective temperature, surface gravity, micro-turbulence velocity, and abundance of elements, such as helium and silicon. In addition, the values of the line force parameters were obtained. The developed algorithm was tested with 40 observed spectra, achieving <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>85</mn><mo>%</mo></mrow></semantics></math></inline-formula> of the expected results according to the scientific literature. The execution time ranged from 6 to 13 min per spectrum, which represents less than <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>5</mn><mo>%</mo></mrow></semantics></math></inline-formula> of the total time required for a one-to-one comparison search under the same conditions.
format Article
id doaj-art-4fa9f4085dba467faeee1e15fa171b3e
institution OA Journals
issn 2227-7390
language English
publishDate 2024-10-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-4fa9f4085dba467faeee1e15fa171b3e2025-08-20T02:10:56ZengMDPI AGMathematics2227-73902024-10-011220316910.3390/math12203169Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning TechniquesEsteban Olivares0Michel Curé1Ignacio Araya2Ernesto Fabregas3Catalina Arcos4Natalia Machuca5Gonzalo Farias6Escuela de Ingeniería Eléctrica, Pontificia Universidad Católica de Valparaíso, Av. Brasil 2147, Valparaíso 2362804, ChileInstituto de Física y Astronomía, Universidad de Valparaíso, Av. Gran Bretaña 1111, Valparaíso 2362804, ChileCentro Multidisciplinario de Física, Universidad Mayor, Santiago 8580745, ChileDepartamento de Informática y Automática, Universidad Nacional de Educación a Distancia, Juan del Rosal 16, 28040 Madrid, SpainInstituto de Física y Astronomía, Universidad de Valparaíso, Av. Gran Bretaña 1111, Valparaíso 2362804, ChileInstituto de Física y Astronomía, Universidad de Valparaíso, Av. Gran Bretaña 1111, Valparaíso 2362804, ChileEscuela de Ingeniería Eléctrica, Pontificia Universidad Católica de Valparaíso, Av. Brasil 2147, Valparaíso 2362804, ChileThis article presents a new algorithm that uses techniques from the field of artificial intelligence to automatically estimate the physical parameters of massive stars from a grid of stellar spectral models. This is the first grid to consider hydrodynamic solutions for stellar winds and radiative transport, containing more than 573 thousand synthetic spectra. The methodology involves grouping spectral models using deep learning and clustering techniques. The goal is to delineate the search regions and differentiate the “species” of spectra based on the shapes of the spectral line profiles. Synthetic spectra close to an observed stellar spectrum are selected using deep learning and unsupervised clustering algorithms. As a result, for each spectrum, we found the effective temperature, surface gravity, micro-turbulence velocity, and abundance of elements, such as helium and silicon. In addition, the values of the line force parameters were obtained. The developed algorithm was tested with 40 observed spectra, achieving <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>85</mn><mo>%</mo></mrow></semantics></math></inline-formula> of the expected results according to the scientific literature. The execution time ranged from 6 to 13 min per spectrum, which represents less than <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>5</mn><mo>%</mo></mrow></semantics></math></inline-formula> of the total time required for a one-to-one comparison search under the same conditions.https://www.mdpi.com/2227-7390/12/20/3169data analysisdeep learningmassive starsfundamental parametersastronomical databases miscellaneous
spellingShingle Esteban Olivares
Michel Curé
Ignacio Araya
Ernesto Fabregas
Catalina Arcos
Natalia Machuca
Gonzalo Farias
Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques
Mathematics
data analysis
deep learning
massive stars
fundamental parameters
astronomical databases miscellaneous
title Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques
title_full Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques
title_fullStr Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques
title_full_unstemmed Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques
title_short Estimation of Physical Stellar Parameters from Spectral Models Using Deep Learning Techniques
title_sort estimation of physical stellar parameters from spectral models using deep learning techniques
topic data analysis
deep learning
massive stars
fundamental parameters
astronomical databases miscellaneous
url https://www.mdpi.com/2227-7390/12/20/3169
work_keys_str_mv AT estebanolivares estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques
AT michelcure estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques
AT ignacioaraya estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques
AT ernestofabregas estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques
AT catalinaarcos estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques
AT nataliamachuca estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques
AT gonzalofarias estimationofphysicalstellarparametersfromspectralmodelsusingdeeplearningtechniques