Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/132827
Title: Detección de tránsitos de exoplanetas mediante técnicas de deep learning
Author: Casal Argüelles, Alejandro
Director: Casas-Roma, Jordi  
Tutor: Ruiz Dern, Laura  
Abstract: The aim of this work is to prepare automatic algorithms, with deep learning techniques, to detect exoplanets from the data collected by NASA's K2 mission. This mission, heir to the Kepler mission, retrieved data on the luminosity of a multitude of stars over time, in what are called light curves. Decreases in apparent luminosity could indicate a planetary transit in front of the star, which it would partially obscure. This is one of the most successful current methods for exoplanet detection. Due to equipment problems, the K2 mission only collected data for periods of about 80 days, in different sectors of the sky, unlike the original Kepler mission, which collected data for several years from a single sector. Thus, the analysis of the K2 mission data presents a major difficulty as it is practically impossible to record the same planetary transit several times. The classical treatments for determining the presence of exoplanets are based on an initial preprocessing of the light curves to detect temporary decreases in luminosity, i.e. possible transits, and their subsequent analysis, as input to predictive models, to determine whether these decreases in luminosity are associated with an exoplanet or not. In this work we propose to use all the data from each light curve, without the aforementioned pre-processing and extraction of possible transits, in an end-to-end approach. This implies an added difficulty, but on the other hand it should allow an earlier analysis of the information that may be available, for example by extrapolating the methods to data from ongoing missions, such as TESS.
Keywords: exoplanet detection
deep learning
Kepler-K2 mission
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 4-Jun-2021
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
acasalTFM0621memoria.pdfMemoria del TFM4,16 MBAdobe PDFThumbnail
View/Open