Predicting the thermal conductivity enhancement of nanofluids using computational intelligence

Nanofluids, composed of nanoparticles in base liquids, have drawn increasing attention in heat transfer applications due to their anomalously increased thermal conductivity. Pertinent parameters, including the base liquid thermal conductivity, particle thermal conductivity, particle size, particle volume fraction, and temperature, have been shown to have significant but complex effects on thermal performance of nanofluids, which is commonly characterized by the thermal conductivity enhancement, E%. In this work, machine learning is used to develop the Gaussian process regression model to find statistical correlations between E% and aforementioned physical parameters among various types of nanofluids. Nearly 300 nanofluid samples, dispersions of metal and ceramic nanoparticles in water, ethylene glycol, and transformer oil, are explored for this purpose. The modeling approach demonstrates a high degree of accuracy and stability, contributing to efficient and low-cost estimations of E%.