A novel Bayesian optimization for flow condensation enhancement using nanorefrigerant: A combined analytical and experimental study

According to the recent researches, adding nanomaterial within the pure refrigerant can substantially enhance the heat transfer rate in phase change (boiling/condensing) flows. Therefore, for a given operating conditions, it is unclear whether the heat transfer is optimized at a certain mass fraction of nanoparticles. In the present study, an integrated analytical, and experimental approach was scrutinized to find the heat transfer optimization of flow condensation using nanorefrigerant. For the experiment, CuO nanoparticles were disperesed with varying mass fractions (0.5–3.5%) in the baseline refrigerant/oil (R600a/POE) to produce Nanorefrigerants (R600a/POE/CuO). The optimum nanomaterial concentration for maximum perfromance has been desiganted by a novel optimization method called two stage-Bayesian optimization (TS-BO), which replaces the expensive experimental process by a cheap surrogate model constructed by Kriging. It was proved that the optimum nanoparticle fraction is significantly depend on the mass velocity and vapor quality while at a fixed vapor quality, the optimum nanoparticle concentration increased with decreasing mass flux. The highest heat transfer enhancement was achieved by nanparticle concentrations of 1.5–2.2%.