Machine learning investigation with neural network modelling for Sutterby Multi-hybrid fluid in biomedical treatments
Hamid Qureshi, Zahoor Shah, Waqar Azeem Khan, Abaker A. Hassaballa, Ahmed Shatat, Taseer Muhammad
Abstract
• Sutterby multi-hybrid fluid is considered here. • Machine learning based technique is considered here. • Levenberg Marquardt neural networks is analyzed. This work presents a new way for estimating thermal conduction properties and other thermal characteristics of Sutterby tetra hybrid nanofluids by leveraging the machine learning expedition with neural network modelling. The nanofluid is a colloidal suspension of copper, silver, iron oxide, and zinc in an Ethanol base fluid. Using similarity transformations, the governing PDEs transform into an ODE system and are solved by Artificial Intelligence (AI)-based techniques via neural networks optimized with Levenberg Marquardt Algorithm for efficient training. The present work mainly focuses on the enhancement in thermal conductivity using tetra-HNF for various engineering applications. The enhanced thermal properties of tetra-HNFs make these nanofluids potential candidates for use in solar energy systems, heat exchangers, and cooling systems. the study employs machine learning and numerical techniques to model and predict thermal conductivity, with results validated with data through error plots. The obtained results prove the feasibility of tetra-HNF to act as effective thermal enhancement materials, while their possible application establishes a future building block for the development of heat transfer technologies and high-efficient energy systems.