Journal Menu




Related Events

Applied Science and Technology Express

ASTE 2022, 13 pages - Article ID: ASTE-2202242112437



Estimation of Dynamic Viscosity of Nanofluids via Graph Neural Network for Enhanced Oil Recovery

Authors

Mohammad Redzuan Firdaus bin Fozi


Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, Johor Bahru, Malaysia.
ABSTRACT

The Graph Neural Network (GNN) is a sort of Artificial Neural Network used to anticipate a system’s performance. In order to calculate the dynamic viscosity of nanofluid, a neural network is usually utilised for simulation. GNN has been proven to be quite beneficial in numerous studies. GraphSAGE and Artificial Neural Network-Multilayer Perceptron are two different forms of GNNs employed in this study (ANN-MLP). To compute the flow of nanofluids in porous media, the ANN-MLP with Levenberg-Marquardt training method is utilised. The model’s data was gathered from the literature. By examining the value of R2, the hidden neurons for the neural network are carefully picked. The GNN is capable of simulating nanofluid dynamic viscosity. The simulation provides beneficial information.

Graph Neural Network; Levenberg-Marquardt; Dynamic Viscosity; Nanofluid.
PDF [ 570.97 Kb ] | Endnote File | XML