Eventos Anais de eventos
COBEM 2019
25th International Congress of Mechanical Engineering
REGRESSION MODELS AND NEURAL NETWORKS IN ESTIMATION OF THE WORKING FLUID TEMPERATURE OF PARABOLIC SOLAR CONCENTRATORS
Submission Author:
Lino Wagner Castelo Branco Portela , PI , Brazil
Co-Authors:
Lino Wagner Castelo Branco Portela, Erilson Barbosa, Ana Almeida
Presenter: Lino Wagner Castelo Branco Portela
doi://10.26678/ABCM.COBEM2019.COB2019-1821
Abstract
This work proposes the verification of the application of the mathematical models Simple Linear Regression, Multiple Linear Regression and Artificial Neural Networks (ANNs) in the prediction of the working fluid outlet temperature a parabolic solar concentrator (CSP) with tracking system. The concentrator is aided by sensors in the measurement of temperature and solar radiation. With the R programming language, it was possible to analyze the collected data and perform statistical calculations, where the Artificial Neural Networks model presented 5.23 in the Root Mean Square Error (RMSE) and 78.89 % in the R² statistic.
Keywords
Simple Linear Regression, Multiple Linear Regression, Artificial neural networks, Parabolic Solar Concentrator, RMSE
