Artificial intelligence and numerical study of the heat transfer and entropy generation analysis of NEPCM-MWCNTs-Water Hybrid Nanofluids inside a quadrilateral enclosure

Abstract

In the present investigation, the primary objective is to assess the synergistic impact of a novel hybrid nanofluid composed of nano-enhanced phase change material, multi-walled carbon nanotubes, and water. The governing equations are transformed into dimensionless forms for a more generalized analysis. To solve the problem, the Galerkin finite element method is employed, offering a robust numerical approach. A dataset of 1000 records was created by numerically solving the model for various combinations of control parameters. Using the dataset, a neural network was trained to learn the relationship between control parameters and heat transfer rate. The evaluation outcomes are comprehensively illustrated through key parameters, including local and average Nusselt numbers, total entropy generation, contours, and streamlines in the range of 103<Rayleigh number<105, 0<nanotube concentration<0.025, 0<nano capsule concentration <0.025, and 0.1<non-dimensional fusion temperature <0.7. Remarkably, the results indicate a substantial improvement in the heat transfer rate of the suspension for a hybrid concentration of 5 %, showcasing an impressive 13 % enhancement in contrast to the water host fluid's performance. Notably, the observed rise in entropy generation is relatively moderate, only a 5 % increase.

Keywords

• Entropy
• Dense Neural Networks
• Hybrid nanofluid
• Nano-encapsulated phase change material (NEPCM)
• Multi-walled carbon nanotubes (MWCNTs)