Numerical and Theoretical Investigation of Thermal Performance in Double Pipe Heat Exchanger Using Nanofluids with Integrated CFD and ANN Methods

Abstract Recent research has focused significantly on the rising energy demand and advancements in heat exchangers, which are effective in heat transfer. This study examines the counter-flow Double Pipe Heat Exchanger (DPHE) under various flow, geometric, and thermal conditions using water-based nanofluids via Computational Fluid Dynamics (CFD). The objective of this research is to investigate the heat transfer performance in DPHEs, employing the effectiveness-NTU method for theoretical validation. Additionally, this study incorporates widely used water-based nanofluids with Al 2 O 3 , CuO, and hexagonal Boron Nitride (h-BN) nanoparticles at varying volume fractions to improve heat transfer. Several cases are designed to assess and compare thermal performance in heat exchangers using CFD. An Artificial Neural Network (ANN) model is developed in Python, integrating MATLAB to estimate thermal performance using the Overall Heat Transfer Coefficient ( U ). The findings revealed that h-BN nanofluids demonstrated outstanding heat transfer performance in different nanoparticles used in this study. The h-BN showed superior thermal performance compared to water and the other water-based Al 2 O 3 and CuO nanofluids. Comparisons with theoretical calculations validated the precision of the CFD simulation outcomes. Under the same conditions, it was observed that Al 2 O 3 increased U values ​​by up to 20%, CuO by up to 15%, and h-BN by up to two times compared to water samples. The effectiveness method is in good agreement with CFD results. The constructed ANN model predicts heat transfer performance with an accuracy of 98.3% when compared with numerical and theoretical results.