“Sistema de alerta temprana de niveles de radiación solar empleando modelo predictivo en la UTEQ”

The present research was carried out with the objective of developing an early warning system for solar radiation levels at the Universidad Técnica Estatal de Quevedo, Manuel Haz Álvarez campus, through the application of predictive modeling techniques. In the development of the predictive model, the relevant meteorological variables that influenced the solar radiation received in the area were identified, that is, the time of day, time of year and temperature. Different regression techniques were used, including linear regression, polynomial regression, decision tree, support vector with linear kernel, support vector with RBF kernel, random forest and gradient boosting, to make effective predictions. of irradiation levels in the web application for predicting solar radiation levels at UTEQ. The predictive model was evaluated by comparing the predicted levels with the actual levels measured by the pyranometer, using evaluation metrics such as mean square error (MSE) and coefficient of determination (R2) to measure the accuracy and fit of the model. The results obtained demonstrate the effectiveness of the predictive model in estimating solar radiation levels in the UTEQ and its usefulness in alerting users in case of dangerous levels.

• Hovedforfatter: Anchundia Troncoso , Jordy Ronaldo (author)
• Format: bachelorThesis
• Sprog: spa
• Udgivet: 2023
• Fag: Variables meteorológicas; Regresión, radiación solar
• Online adgang: https://repositorio.uteq.edu.ec/handle/43000/7127