Thermophysical properties using ND/water nanofluids: An experimental study, ANFIS-based model and optimization

The research work was achieved to optimize the experimentally determined thermophysical properties of water based nanodiamond nanofluids using the Adaptive network-based fuzzy inference system model. The thermophysical properties were determined at various temperatures (20 °C to 60 °C) and various particle volume concentrations (0.2% to 1.0%). The stability of nanodiamond nanofluids was measured based on the dynamic light scattering method. Results indicated that the zeta potential of all the prepared nanodiamond/water nanofluids is above −30 mV. The thermophysical properties like thermal conductivity and viscosity augments are 22.86% and 79.16% at ϕ = 1.0 vol% compared to the water data at a temperature of 60 °C. Fuzzy logic is one of the artificial intelligence tools that have been used to analyze the optimized property value. The optimization process was applied in a single objective and a multi-objective procedure using a new and efficient optimizer, namely, the marine predators' algorithm. In a single objective, the lowest density and highest thermal conductivity values were found at 0.23% with 54.81^o C and 1% with 60 °C, respectively. However, the viscosity and specific heat showed improvements only in the absence of the nanodiamond material and temperatures of 35.99 and 60 °C, respectively. In the multi-objective process, the optimizer confirmed that the thermophysical properties' optimal values could be obtained when no nanomaterial was added, and the best values were found at a temperature of 59.48 °C. This optimal point is found as close to the experimental data.