Bow-shaped vortex generators in finned-tube heat exchangers; ANN/GA-based hydrothermal/structural optimization

Abstract

Temperature plays a crucial role in various industries, equipment, and human life, and heat exchangers emerge as fundamental devices in managing this essential parameter. Heat exchangers come in various types, each tailored for specific applications. One commonly utilized variant is the finned-tube heat exchanger, which excels in transferring heat between liquid and gas. This study involved a numerical investigation of a finned-tube heat exchanger equipped with four bow-shaped vortex generators. The primary objective of this study was to assess the influence of three distinct geometric parameters associated with vortex generators on the performance of the heat exchanger. The key performance indicators investigated in this research were the Nusselt number and friction factor, which respectively represent the thermal and hydraulic performance of the device. Also, the relative efficiency index parameter was investigated to evaluate thermal and hydraulic performances simultaneously. Artificial Neural Network and Genetic Algorithm methods were employed to develop response models to simplify and increase the analysis accuracy and determine the optimal geometry. The artificial neural network's prediction models demonstrated high accuracy and effectively forecasted the desired outcomes. It was determined that the relative efficiency index of the finned-tube heat exchanger with vortex generators, even in its weakest-performing design, reached 1.0467. This means that by adding the bow-shaped vortex generators to the heat exchanger, there is a 4.67% increase in the overall performance in the worst case. Also, after predicting the optimal design for the heat exchanger, it was observed that this device has a 13.12% improvement in overall performance compared to the case without vortex generators. To adjust the Nusselt number of the finned-tube heat exchanger lacking vortex generators with that of the finned-tube heat exchanger featuring the optimal design, it is necessary to increase the inlet flow rate of the vortex generators-lacking heat exchanger by 53.20%. This increase in flow rate leads to a significant escalation of 108.73% in the heat exchanger's pumping power compared to the pumping power of the optimal design heat exchanger. © 2024 The Authors

Keywords Artificial Neural Network, Heat transfer improvement, Numerical Model, Nusselt number, Optimization