Investigation of convective heat transfer of ferrofluid using CFD simulation and adaptive neuro-fuzzy inference system optimized with particle swarm optimization algorithm

Ferrofluid is defined as a magnetic fluid which is composed of magnetic nanoparticles immersed in the base fluid such as water and oil. Nanofluids under magnetic field were proposed as a novel working fluid for industrial applications. In this study, the convective heat transfer of Fe3O4/water ferrofluid under constant magnetic field is evaluated. For this purpose, computational fluid dynamics (CFD) simulation and adaptive neuro-fuzzy inference system optimized with particle swarm optimization (ANFIS-PSO) are applied. To develop the ANFIS-PSO model, inlet temperature of ferrofluid, volume fraction of nanoparticle (Fe3O4), Reynolds number and intensity of magnetic field are considered as input variables of the model and heat transfer coefficient (HTC) of Fe3O4/water ferrofluid is considered to be the target. The results demonstrated that the developed ANFIS-PSO model can successfully predict the HTC of ferrofluid in laminar and turbulent flows in terms of the correlation coefficient (R), root mean square error (RMSE) and mean absolute percentage error (MAPE) respectively with 0.9992, 117.19 (W/m(2)K) and 2.44% for testing phase of the network. Also, CFD simulation and ANFIS-PSO model illustrated that the amount of the HTC of ferrofluid increases by increasing in intensity of magnetic field and inlet temperature of ferrofluid. (C) 2017 Elsevier B.V. All rights reserved.