Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/150503
Title: Diseño, desarrollo y evaluación de modelos de predicción de series temporales aplicados al monitoreo geotécnico y estructural
Author: Marin Fidalgo, Juan  
Tutor: Sanchez, Friman  
Others: Isern, David  
Abstract: While predictive techniques are being applied in other fields of knowledge, they are not being utilized in geotechnical and structural monitoring. However, sampling and data acquisition technologies have evolved, and there is now a vast availability of data that can be analyzed using advanced computational learning techniques. This work focuses on the analysis, design, implementation and evaluation of time series prediction techniques applied to geotechnical and structural monitoring. It explores the effectiveness of a variety of time series analysis models, from classical statistical methods to neural networks. Additionally, a platform for visualization and analysis of the models has been implemented with a modular architecture that adapts to the various present technologies. Monitoring data are characterized by being non-stationary and presenting complex seasonality, as well as non-linear relationships between the components of the series. This behavior is more pronounced in structural sensors. These challenges make traditional statistical models, which assume linear relationships and have difficulty modeling complex seasonality, inadequate. In contrast, recurrent neural networks show superior performance in predicting time series in this context.
Keywords: predicción de series temporales
aprendizaje automático
redes neuronales recurrentes
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: Jun-2024
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 
TFG_PRESENTACION_JMF.pptxPresentación19,69 MBMicrosoft Powerpoint XMLView/Open

TFG_PRESENTACION_JMF.mp4

Exposición518,35 MBMP4View/Open
jmarinfiTFG0624memoria.pdfMemoria del TFG16,46 MBAdobe PDFThumbnail
View/Open
jmarinfiTFG0624presentacion.pdfPresentación en PDF del TFG31,07 MBAdobe PDFThumbnail
View/Open
Share:
Export:
View statistics

This item is licensed under aCreative Commons License Creative Commons