Artificial neural network for prediction of thermal conductivity of rGO–metal oxide nanocomposite-based nanofluids

• Published in: Neural computing & applications 2022, Vol.34 (1), p.271-282
• Main Authors: Barai, Divya P., Bhanvase, Bharat A., Pandharipande, Shekhar L.
• Format: Article
• Language: English
• Subjects: Artificial Intelligence; Artificial neural networks; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer Science; Data Mining and Knowledge Discovery; Data points; Heat conductivity; Heat transfer; Image Processing and Computer Vision; Metal oxides; Model testing; Nanocomposites; Nanofluids; Neural networks; Original Article; Probability and Statistics in Computer Science; Thermal conductivity
• ISSN: 0941-0643; 1433-3058
• DOI: 10.1007/s00521-021-06366-z

A four-input artificial neural network (ANN) model has been presented for the prediction of thermal conductivity of rGO–metal oxide nanocomposite-based nanofluids. For this, data of five types of water-based nanofluids containing rGO–metal oxide nanocomposites particles were used from the available literature. The four-input variables considered were molecular weight of nanocomposite, average particle size of nanocomposites, concentration, and temperature of nanofluid which exhibited thermal conductivity of the nanofluids as output. Using the same architecture, two ANN models were developed, one using a total of 185 data points and the other by dividing the data points in two sets (training and testing). The model agreed well with the experimental data and exhibited an R 2 value of 0.956 for the testing data set. Also, the magnitude of deviation of the predicted thermal conductivity for all the data points was very less with an average residual of ± 0.048 W/mK.