Application of Artificial Neural Network (ANN) for the prediction of thermal conductivity of oxide–water nanofluids

A review on the heat and mass transfer phenomena in nanofluid coolants with special focus on automotive applications

“…Finally, theoretical models of nanofluid viscosity may also account for particle aggregation [201,236,240].…”

Experimental study on thermal conductivity of DWCNT-ZnO/water-EG nanofluids

“…Information of 34 the rheological behavior of nanofluids is found to be very important 35 in determining their appropriateness for heat transfer applications. 36 Hence, many researchers around the world have investigated the behav- 37 iors of nanofluids. A summary of existing experimental studies for 38 the thermal conductivity of different oxide nanofluids is presented in 39 Table 1.…”

“…Then, using neural network, they modeled the 62 experimental results using temperature and volume fraction of nano-63 particles as input, and thermal conductivity of nanoparticles as output 64 of neural network. Longo et al [37] proposed two artificial neural net-65 work models for estimating the thermal conductivity of Al 2 O 3 /water 66 and TiO 2 /water nanofluids by the temperature, volume fraction, diame-67 ter of nanoparticle and particle thermal conductivity as the input Hemmat Esfe et al [38] using experimental data. In their study, the …”

Thermal conductivity of Cu/TiO2–water/EG hybrid nanofluid: Experimental data and modeling using artificial neural network and correlation

“…Furthermore, in another investigation and based on the experimental data, they proposed two correlations that show the relationships between viscosity, solid concentration, and the nanofluid's temperature [14]. Longo et al [15] used three-input and four-input artificial neural networks to predict the thermal conductivity ratio of oxide-water nanofluid. They employed the temperature, solid concentration, and thermal conductivity ratio of the nanofluid as the three inputs for both of the networks, with the effect of the nanoparticle clusters' average size also employed for the four-input network.…”