Thermal performance of hybrid fly ash and copper nanofluid in various mixture ratios: Experimental investigation and application of a modern ensemble machine learning approach

The purpose of the current research work is to investigate the various properties like thermal conductivity, stability and viscosity of copper and a mixture of fly ash-Copper (FA:Cu) nanoparticles (NP) suspended in water. The research experiments were conducted for a concentration of 1.0 vol% of FA:Cu hybrid nanofluid (HNF) with various mixture ratios. The measurements of thermal conductivity and viscosity were performed in the 30-60 degrees C temperature range. The highest thermal conductivity and viscosity values for HNF with a mixture ratio of 20:80 were obtained with a maximum amplification exceeding 83.2% and 65% than the base fluid, respectively. The properties enhancement ratio (PER) reveals that HNF with 1.0 vol% concentration enhances heat transfer for all defined mixture ratios in the reported research work. Finally, the predictability potential of an ensemble-based machine learning technique called Boosted Regression Tree (BRT), based on nanofluids temperature (T) and mixture ratio (R), were compared with the classical regression approach to simulate the thermo-physical properties of HNF. The outcomes of the BRT model in terms of (r(viscosity) = 0.9953 and r(thermal conductivity) = 0.9991) superior to the regression method with (r(viscosity) = 0.9695 and r(thermal conductivity) = 0.9539) over the viscosity and thermal conductivity prediction process.

Automated summary

The research aims to investigate the properties of copper and fly ash-copper nanoparticles suspended in water, with the highest thermal conductivity and viscosity values obtained at a mixture ratio of 20:80 for HNF. The study shows that HNF with a concentration of 1.0 vol% enhances heat transfer for all mixture ratios. Additionally, the Boosted Regression Tree model outperforms classical regression in predicting the thermo-physical properties of HNF.