Analysis of Swirling Flow of Hybrid Nanofluids Due to a Moving Cylinder With Temperature‐Dependent Thermal Conductivity Through Machine Learning Algorithms

ABSTRACT The current work is motivated to study the flow characteristics and transfer of heat in the occurrence of tiny‐sized solid particles, which are utilized to improve the thermo‐physical features of the base fluid through machine learning. In this work, the significant consequences of viscous dissipation are due to a swirling flow of hybrid nano‐fluid on a stretched cylinder. The non‐Newtonian working medium's effect on the flow structure, heat transfer, and entropy generation performance is investigated by means of partial differential governing equations. The thermal radiation impacts and temperature‐dependent variable conductivity are also characterized. Swirling flow and slip impacts are considered in this study. The base fluid (ethylene glycol with nanoparticles of copper () and titanium dioxide () is adopted in this communication. The boundary flow equations formulated in the present model are converted into ODEs through the application of appropriately modified variables. Furthermore, two distinct numerical techniques are applied to solve the system of ODEs derived from modeled PDEs. A bvp4c algorithm built‐in computational technique in MATLAB is used to numerically calculate the dimensionless nonlinear ODEs. For comparative analysis, artificial intelligence based neural network back propagation Levenberg Marquardth scheme is used. The two methods yielded matching numerical and visual outcomes, confirming the strong consistency and dependability of the computational system. The physical inspirations of prominent parameters against the velocity profile and temperature distribution are analyzed through graphs and tabular data.