Skip to content
Open access

Nanoparticle Size Effects on Thermal Conductivity of Hybrid Nanofluids: Experiments and Neural Network Modeling

Oct 2026 · Crystals · 0 citations · 52 references

Abstract

Traditional cutting fluids are insufficient to mitigate problems such as the high temperatures and tool wear that occur during machining processes. Therefore, the use of nanoparticle-modified coolant oils has gained importance in recent years. In this study, hybrid nanofluids were prepared using different nanoparticles (hBN, ZnO, MWCNT, TiO2, Al2O3), and a multilayer perceptron neural network (MLPNN)-based model considering particle size was developed to predict the thermal conductivity of these fluids. The dataset consists of 180 data points measured in the lab for each single nanofluid and 5 different hybrid nanofluids. The model results show high accuracy (R2 = 0.9997). Error metrics were calculated as MAE = 0.001875, MSE = 7.61 × 10−6, and MAPE = 0.1243. Residual analyses showed that the model did not generate systematic errors and could make reliable predictions for different hybrid nanoparticle combinations. Compared to studies in the literature, the proposed model performed competitively and, in most cases, better in terms of error metrics. In conclusion, the proposed model, taking particle size into account, offers a fast, reliable, and practical method for estimating the thermal conductivity of nanoparticle-modified coolant oils.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.