An artificial intelligence – Finite element study of magnetohydrodynamic heat transfer of a nano-encapsulated phase change material suspension in a cylindrical enclosure with partial heated walls

• Published in: International communications in heat and mass transfer Vol. 164; p. 108812
• Main Authors: Aljibori, Hakim S. Sultan, Ali, Mohammed Hasan, Raizah, Zehba, Hajjar, Ahmad, Shafi, Jana, Alresheedi, Faisal, Elhassanein, Ahmed, Edalatifar, Mohammad, Ghalambaz, Mohammad
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2025
• Subjects: Artificial intelligence; Dataset; Entropy; Nano-encapsulated phase change material suspension; Phase transition region; Point source magnetic field; Point source magnetic field; Entropy; Nano-encapsulated phase change material suspension; Artificial intelligence; Dataset; Phase transition region
• ISSN: 0735-1933
• DOI: 10.1016/j.icheatmasstransfer.2025.108812

The present study delivers an exhaustive exploration into the mechanisms governing heat and mass transfer within an enclosure filled with Nano-Encapsulated Phase Change Material (NEPCM) and subject to magnetic fields. Leveraging a set of governing partial differential equations, the research investigates the intricacies of natural convection, convective heat transfer, and conjugate heat transfer within solid walls. The entropy generation in the NEPCM suspension was addressed. The finite element method was used for the numerical solution. A dataset was generated from simulations and was utilized to train an artificial neural network. Subsequently, the trained neural network was employed to analyze the effects of model parameters on heat transfer and entropy generation. The results show the pivotal role of the magnetic source's position in manipulating heat transfer. Increasing the magnetic number slightly elevates NuAvg and significantly boosts total entropy generation. An increment in thermal conductivity ratio from 1 to 100 at Ra = 100,000 elevates NuAvg from 2 to 4.5, signifying a 56 % improvement. Additionally, elevating nanoparticle concentration enhances NuL, owing to increased thermal conductivity. The study reveals that regions of high entropy generation are universally close to the magnetic source, regardless of its specific location.