Eventos Anais de eventos
ENCIT 2020
18th Brazilian Congress of Thermal Sciences and Engineering
Case study of the metamodels of a shell-and-tube heat exchanger
Submission Author:
Felipe Raul Ponce Arrieta , MG
Co-Authors:
Wagner Henrique Saldanha, Felipe Raul Ponce Arrieta, Gustavo Soares
Presenter: Felipe Raul Ponce Arrieta
doi://10.26678/ABCM.ENCIT2020.CIT20-0449
Abstract
This paper is aimed at a case study in which a shell-and-tube heat exchanger (STHE) metamodel is developed from its analytical functions. The techniques for generating the STHE metamodel were Multivariate Adaptive Regression Splines (MARS) and the Multilayer Perceptrons (MLP) neural network, being implemented in the toolbox MATLAB ARESLab and nnstart, respectively. The results obtained by the two techniques had suitable Mean Square Error (MSE), thus, it is concluded that these techniques are suitable for the development of the STHE metamodel, and should continue to be explored in other engineering problems.
Keywords
Metamodel, Shell-and-Tube Heat Exchanger, Multivariate Adaptive Regression Spline, Neural Network
DOWNLOAD PDF VIEW PRESENTATION
