Heat transfer enhancement in vehicle cabin heating system with hybrid nanofluid: An experimental and artificial intelligence approach

Abstract

In louvered fin-and-flat-tube aluminum heat exchangers, the interrupted surfaces of the louvered fin channel, through which air flows, enhance heat transfer by growing and degrading laminar boundary layers, while the large surface-to-cross-section flow area ratio of the flat tube further enhances heat transfer. Consequently, due to their relatively good heat transfer performance and compact design, louvered fin-and-flat-tube aluminum heat exchangers are often utilized for cabin heating systems. Considering previous studies involving louvered fin-and-flat-tube aluminum heat exchangers with conventional fluids and with mono-nanofluids, this is the first study employing Al₂O₃-TiO₂/water hybrid nanofluid experimentally in this context, and utilizing a forward-looking ANN forecasting model for scenarios that cannot be experimentally conducted in such systems. Accordingly, the heat transfer enhancement in the louvered fin-and-flat-tube heat exchanger is experimentally investigated for both pure water alone and for different volume concentrations of Al₂O₃-TiO₂/water hybrid nanofluid. Heat transfer rate, convection heat transfer coefficient, Nusselt number, effectiveness, overall heat transfer coefficient, pressure drop, and performance evaluation index were assessed for different inlet temperatures (50 °C, 60 °C, 70 °C) and volume flow rates (3.65 LPM, 5.65 LPM, 7.65 LPM) in the range 0–0.2 % volume concentration. The experimental results show that the highest heat transfer rate enhancement was 113.32 % using the hybrid nanofluid at φ = 0.2 % compared to pure water at a 50 °C inlet temperature and 7.65 LPM flow rate. It was further shown that even using only the hybrid nanofluid, the effectiveness can be increased up to 0.422 at a constant fan speed without the need to enlarge the heat exchanger size, at a flow rate of 7.65 LPM and an inlet temperature of 70 °C. It has also been found that the maximum values of the performance evaluation index calculated according to two different formulas are 2.260 and 2.049 at φ = 0.2 %. In the following part of the study, heat transfer rate, effectiveness, and pressure drop values in the instances of 0.3 % and 0.4 % volume concentrations were forecast using an artificial neural network model trained with the Levenberg-Marquardt algorithm using experimental data. With the proposed artificial neural network model, the forecast heat transfer rates were as expected, while the expected trend could not be achieved in terms of the effectiveness and pressure drop predictions. The study ultimately found that the utilization of a Al₂O₃-TiO₂/water hybrid nanofluid in cabin heating systems, even at a relatively low concentration value of 0.2 %, is capable of achieving very high heat transfer enhancement in exchange for only a very minor pressure drop.