Application of deep learning techniques in modelling and observation of the solar photosphere

This work is part of the applications of neural networks in the study and modeling of the phenomena presentin the solar photosphere. The proposed research is based on the network model generative adversaries usingPytorch’s artificial intelligence modules. We aim at training a neural network capable of generating groupsof images of a high similarity with input images, These images correspond to physical magnitudes of thesolar photosphere such as density, field magnetic field, plasma velocity, temperature, among others, obtainedfrom the MURaM simulation code, although the neural network can be trained to generate images of anyphysical magnitude. The work is focused on the generation of magnetic field images in the solar photosphere.Results of the neural network training process are presented, as well as the comparison between the trainingand generated images, and the challenges to use these tools in the study of the solar photosphere.

• Fecha de publicación: 2022
• Institución: Universidad Pedagógica y Tecnológica de Colombia
• Repositorio: RiUPTC: Repositorio Institucional UPTC
• Idioma: spa
• OAI Identifier: oai:repositorio.uptc.edu.co:001/15377
• Acceso en línea: https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/15240; https://repositorio.uptc.edu.co/handle/001/15377
• Palabra clave: GAN, DCGAN, Pytorch, fotósfera.; GAN, DCGAN, Pytorch, photosphere.
• License: http://purl.org/coar/access_right/c_abf2