Simulation and Optimization of Finned and Twisted Double‐Tube Heat Exchanger Using Neural Network Algorithm

ABSTRACT Creating a high heat transfer flux is not a common phenomenon and occurs only when a source of heat production (or consumption) is placed in a small volume; in precision casting systems, heat transfer should be done at a high velocity due to the presence of thin molten branches. Nanofluids present a significant opportunity to improve the thermal efficiency. In this research, a combination of different solutions to enhance heat transfer has been evaluated simultaneously. For this purpose, in the current research, aluminum oxide non‐Newtonian nanofluid in volume percentages of 0, 0.5, 1.5, and 2 has been investigated in a torsional heat exchanger with a rotating triangular blade around the tube. The base fluid of this non‐Newtonian nanofluid consists of water with 0.1% mass of carboxymethyl cellulose. In this study's results, the heat exchanger's performance has been predicted. Combining neural network optimization code with the numerical simulation of the double‐tube spiral geometry and using non‐Newtonian nanofluid to increase the heat transfer and improve the performance of the heat exchanger is the innovation of the present research.