ANN-Based Simulation of <i>Cu-TiO</i> ₂ /Water–EG Hybrid Nanofluid for Environmentally Sustainable Cooling in Extreme Temperature Conditions

The increasing frequency of extreme heat events associated with climate change has intensified the demand for sustainable and energy efficient thermal management systems. This study presents a numerical and artificial intelligence assisted investigation of a Maxwell hybrid nanofluid composed of titanium dioxide and copper nanoparticles suspended in a water ethylene glycol base fluid, flowing over a two dimensional stretching surface embedded in a porous medium. The model incorporates magnetic field effects, Hall current, viscous dissipation, binary chemical reactions, and multiple slip boundary conditions. The governing partial differential equations are reduced using similarity transformations and solved numerically using MATLAB’s boundary value problem solver. To enhance computational efficiency, an artificial neural network trained using the Levenberg Marquardt algorithm is developed as a surrogate model for predicting heat flux and transport characteristics. The results reveal that thermal and mass transfer rates are highly sensitive to the Deborah number, magnetic field strength, Eckert number, and Hall parameter. The copper titanium dioxide hybrid nanofluid exhibits superior thermal performance under elevated thermal loads, highlighting its suitability for advanced cooling systems operating in extreme temperature environments. The neural network predictions show strong agreement with the numerical results, demonstrating their potential for rapid and reliable thermal forecasting.