ANN Analysis of Heat Exchanger Design – A numerical Approach


Mercan H. , Tanriverdi U.

ICAME, İstanbul, Turkey, 20 - 22 October 2021, pp.1544

  • Publication Type: Conference Paper / Summary Text
  • City: İstanbul
  • Country: Turkey
  • Page Numbers: pp.1544

Abstract

Heat exchangers have a very broad range of applications. Various
parameters such as working fluid type, inner and outer diameters and length of the
heat exchanger affect the outlet temperature. Due to highly non-linear nature of
those parameters Artificial Neural Networks (ANN) technique is an effective
alternative to construct realistic relations between all these parameters. In this
study, for a heat exchanger where the fin number and geometry is fixed, an
efficient ANN is developed to estimate the outlet temperature as a function of the
size. The comparison of ANN estimate and ANSYS results for overall heat transfer
coefficient and output temperature values are compared. It is observed that the
ANN estimates agree very well with the ANSYS analysis results.