Effectiveness and number of transfer units of plate heat exchanger with Fe3O4–SiO2/Water hybrid nanofluids: Experimental and artificial neural network predictions

Abstract

This paper analyses the effectiveness and number of transfer units of water dispersed Fe3O4–SiO2 hybrid fluids passing via plate heat exchanger experimentally. The flow of Fe3O4–SiO2/water hybrid nanofluids in plate heat exchanger remains the laminar flow conditions. The experimental studies were carried out with particle loadings ranging from 0.2 % to 1.0 % and flow rates ranging from 0.05 kg/s to 0.1166 kg/s. The obtained results of effectiveness and number of transfer units from these experiments were predicted with artificial neural network based scaled conjugate gradient (SCG) method. Based on the chemical reaction and in-situ growth methods, the Fe3O4–SiO2 nanoparticles were prepared and then characterized using x-ray diffraction (XRD) and vibrating sample magnetometer (VSM) equipment. Water diluted hybrid nanofluids were developed and used for the experimental studies. Results from the XRD indicates that, the particle size of synthesized Fe3O4–SiO2 nanoparticles were found to be 47 nm and the results from VSM shows, the final Fe3O4–SiO2 sample contains 67.8 % of SiO2 and 32.2 % of Fe3O4. Results also indicates with an increase of volume loadings and Reynolds number, the effectiveness and number of transfer units were enhanced. The effectiveness and number of transfer units were increased by 13.23 % and 24.51 % at 1.0 % vol. of hybrid nanofluid and at a Reynolds number of 571.89, over water data. Through the SCG method, the correlation coefficient R2 for effectiveness is 0.99040, and the R2 for number of transfer units is 0.98771, respectively. New generalized equation was proposed for the evaluation of effectiveness and number of transfer units.