Performance predictions of laminar heat transfer and pressure drop in an in-line flat tube bundle using an adaptive neuro-fuzzy inference system (ANFIS) model

Abstract

This paper shows how to predict the heat transfer and pressure drop for in-line flat tube configuration in a crossflow, using an adaptive neuro-fuzzy inference system (ANFIS). A numerical study of a 2D steady state and incompressible laminar flow for in-line flat tube configuration in a crossflow is also considered in this study. A finite volume technique and body-fitted coordinate system is used to solve the Navier–Stokes and energy equations. The Reynolds number varies from 10 to 320. Heat transfer and pressure drop results are presented for a tube configuration at transverse pitch and longitudinal pitch. The variation in velocity profile, isotherm contours and streamlines were compared for various configurations. The predicted results for average Nusselt number and dimensionless pressure show a good agreement with available previous work. The accuracy between numerical values and ANFIS model results were obtained with a mean relative error for average Nusselt number, pressure drop less than 1.9% and 2.97% respectively. Therefore, the ANFIS model is capable of predicting the performance of thermal systems in engineering applications, including the model of the tube bundle for heat transfer analysis and pressure drop.