The effect of pH on stability and thermal performance of graphene oxide and copper oxide hybrid nanofluids for heat transfer applications: Application of novel machine learning technique

Abstract

This paper investigates the effects of pH on stability and thermal properties of copper oxide (CuO), graphene oxide (GO), and their hybrid nanofluid (HNF) at different mixing ratios. Initially, sol-gel and Hummer’s method was employed for the synthesis of CuO and GOnanoparticles (NPs), and they are characterized with various techniques. The effects of two different surfactants were analyzed on nanofluid’s (NF’s) stability at different pH values. The properties like thermal conductivity (TC) and viscosity (VST) of NFs were measured at different volume concentration (0.1 vol% to 1.0 vol%) and temperature range of 30–60 °C, respectively. The TC and VST of GO/CuO (50:50) HNF are higher than that of GO/CuO (20:80). The figure of merit (FOM) is determined for the studied HNFs. The correlations were presented to calculate the TC as well as VST of HNFs. Two modern novel machine learning-based ensemble approaches were employed for predictive model development for TC and VST of considered HNFs. The comparison of prognostic models with Taylor’s diagram revealed that Bayesian optimized support vector machine (BoA-SVM) was superior to Bayesian optimized boosted regression trees (BoA-BRT) for both TC and VST models.