Liquid Paraffin Thermal Conductivity with Additives Tungsten Trioxide Nanoparticles: Synthesis and Propose a New Composed Approach of Fuzzy Logic/Artificial Neural Network

A nanofluid (Nf), which contains tungsten trioxide, WO3, nanoparticles as solid-particles distributed in liquid paraffin, is produced. A set-up is provided to experimentally measure the thermal conductivity (TC) of this mixture at various mass fractions and temperatures. Further, two universal approximators of fuzzy logic (FL) and artificial neural network (ANN) are presented to predict the TC of this mixture. The present work's novelty can be presented as: prepare a WO3/paraffin nanofluid, measure it's TC, and develop a novel statistical/numerical model of FL compared with ANN. As it can be seen, temperature and concentration effects positively on nanofluid TC. Moreover, ANN and fuzzy models show suitable precisions; however, the interpolation capability of the proposed fuzzy model outperforms ANN at non-trained inputs.

Automated summary

The study focuses on preparing a nanofluid with WO3 nanoparticles and predicting its thermal conductivity using fuzzy logic and artificial neural network models. The effects of temperature and concentration on the nanofluid's thermal conductivity are positively correlated.