Using neural network and RSM to evaluate improvement in thermal conductivity of nanodiamond-iron oxide/antifreeze

Khetib, Yacine and Alahmadi, Ahmad and Alzaed, Ali and Sajadi, S. Mohammad (2021) Using neural network and RSM to evaluate improvement in thermal conductivity of nanodiamond-iron oxide/antifreeze. Chemical Engineering Communications, 210. pp. 596-606.

[thumbnail of Research Article] Text (Research Article)
Article_CEC_27-9-2021.pdf
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (249kB)

Abstract

This study aimed to investigate the accuracy of artificial neural networks (ANN) in estimating the thermal conductivity (k) of ferrofluid-based nanofluids. The thermal conductivities of kND+Fe2O3/EG-water and kEG-water were measured at temperatures ranging from 20 to 60°C and at volume fractions of 0.05, 0.1, and 0.2%. Results showed that kFe2O3/EG-water was greater than kEG-water by 89%, achieved at 60°C and 0.2% volume fraction.To estimate kND+Fe3O4/EG-water, a three-layer ANN was developed with two, three, and one neuron in its respective layers. This neural network achieved an estimation error of less than 0.8%, yielding a coefficient of determination (R²) of 0.996. Additionally, response surface methodology (RSM) was employed, revealing that cubic polynomial models could predict kND+Fe3O4/EG-water with an error of less than 0.5% and an R² of 0.994

Item Type: Article
Uncontrolled Keywords: ANN, ferrofluid, nanofluid, neuron, RSM.
Subjects: Q Science > Q Science (General)
T Technology > T Technology (General)
Divisions: Department of Communication and Computer Engineering > Research papers
Depositing User: ePrints Depositor
Date Deposited: 08 Oct 2024 10:29
Last Modified: 08 Oct 2024 10:29
URI: https://eprints.cihanuniversity.edu.iq/id/eprint/1620

Actions (login required)

View Item
View Item