Total entropy and exergy efficiency of graphene oxide/nanodiamond hybrid nanofluids in a mini heat sink: experimental and particle swarm optimization predictions

Through the Particle Swarm Optimization (PSO), the thermal entropy, frictional entropy, entropy generation ratio, entropy generation number, and exergy efficiency of reduced graphene oxide/nanodiamond (rGO/ND) hybrid nanofluids flow in a mini-heat sink were predicted after being measured experimentally. A 60:40% (weight percentage) water and ethylene glycol mixture was used as the base fluid in this study. The experiments were conducted under different volume loadings and different Reynolds numbers . Eventually, the thermophysical properties were also estimated. The thermal entropy generation of 2.0% vol. was decreased by 34.87%, and frictional entropy generation and exergy efficiency were raised by 21.30% and 18.10% at a Reynolds number of 4181.24 over the base fluid. The PSO artificial neural network method was used in this study. The PSO predictions data have shown a good acceptance with the experimental values with root-mean-square errors of 0.058262, 4.9088e-05, 0.0034824, 0.015519, and 0.050993, with correlation coefficients of 0.99811, 0.99218, 0.99849, 0.99812, and 0.99571, for thermal entropy, frictional entropy, entropy generation ratio, entropy generation number, and exergy efficiency, respectively. Based on the polynomial regression analysis, new thermal entropy generation, frictional entropy generation, entropy generation number, and exergy efficiency correlations were proposed.