Artificial Neural Networks (ANN) to Predict Overall Heat Transfer Coefficient and Pressure Drop on a Simulated Heat Exchanger

Abstract

Computational Fluid Dynamics (CFD) numerical simulations were performed to calculate the maximum overall heat transfer coefficient (U) and minimum pressure drop (Δp) for a crossflow heat exchanger using four materials: stainless steel, copper, aluminum and titanium. Transversal and longitudinal sections were modified, obtaining 143 geometries for analysis. With the simulated data, an Artificial Neural Network (ANN) was built to predict the overall heat transfer coefficient and pressure drop as a function of the heat exchanger material. The ANN exhibits maximum deviations, between the predicted and simulated data, below 0.9 y 0.3 % for the pressure drop and air overall heat transfer coefficient respectively. This assisted model reference strategy can be used for material selection in the heat exchanger design considering replacement and cleaning cycles due to corrosion and fouling in other thermal analysis tasks in engineering applications.