Thermophysical profile of graphene oxide and MXene hybrid nanofluids for sustainable energy applications: Model prediction with a Bayesian optimized neural network with K-cross fold validation

Abstract

Nanofluids (NFs) as heat transfer fluids (HTFs) have enormous promise in heat exchange systems. One of the key problems for the use of NFs is how to balance effective thermal conductivity (TC), dispersion stability, and viscosity (VST). In this work, graphene oxide (GO) and MXenes nanoparticles (NPs) were synthesized. Then, water base hybrid nanofluids (HNFs) were prepared for a volume concentration of 0.5 % of GO-MXene with various particle mixture ratios. The synthesized NPs were characterized with scanning electron microscopy (SEM), X-ray diffraction (XRD), Fourier transformation spectroscopy (FTIR), and UV–Visible spectroscopy. The specific heat capacity (SHC), VST, and TC of NFs were investigated experimentally at a temperature range of 25 to 60 °C. The pH and stability of NFs were also investigated. The outcomes show that MXene can significantly enhance the dispersion stability of GO NFs. The TC and VST of GO NF are decreases with the addition of MXene. However, the SHC of GO NF is improving with MXene concentration and temperature. The correlations were established for TC, VST, and SHC of developed NFs in the studied range. The figure-of-merit shows that NFs are advantageous for sustainable energy applications like solar energy in the observed concentrations and temperatures. Finally, an intelligent approach, a Bayesian-optimized ANN model for VST, SHC, and TC was developed. The models produced were accurate forecasting models due to their low model errors, strong correlation coefficient (>99 %), and high Kling-Gupta efficiency (>99.15 %).