Exergy Prediction Model of a Double Pipe Heat Exchanger Using Metal Oxide Nanofluids and Twisted Tape Based on the Artificial Neural Network Approach and Experimental Results

In heat transfer area, researches have been carried out over several years for the development of convective heat transfer enhancement (HTE) techniques. For proper optimization of thermal engineering systems in terms of design and operation, not only the heat transfer has to be maximized but also the exegetic efficiency has to be minimized as well. Present study provides a theoretical, numerical, and experimental investigation of the exergy analysis in a double pipe heat exchanger. For this purpose, metal oxide-water nanofluids and twisted tapes (TTs) are considered as the model fluids and turbulators. Results are verified with well-known correlations. The results show that nanofluids and TTs can increase the exergetic efficiency by 30–100% compared to empty tube and water as a base fluid. In addition, the exergetic efficiency increases with increase in nanoparticles concentration and decreases in twist ratio. CuO nanofluid gives better enhancement in exergetic efficiency than others under the same condition. Since the prediction of exergetic efficiency from experimental process is complex and time-consuming process, an ant colony optimization–back propagation (ACO_R–BP) artificial neural networks (ANN) model for identification of the relationship, which may exist between the thermal and flow parameters and exergetic efficiency, have been developed. The network input consists of 11 parameters (⁠$C,nf,Cbf,ρbf,ρnf,ϕ,kbf,knf,μbf,μnf,unf,ubf$⁠) that crucially dominate the heat transfer process. The results indicate that ACO_R–BP ANN provides a high degree of accuracy and reliability. The proposed ANN model can be used to understand how key parameters affect exergetic efficiency without using extensive numerical modeling or experimental studies.