Machine Learning Artificial Neural Network Approach in Predicting Nusselt Number on Micropolar (CuO–Ag–H ₂ O) Fluid Over an Inclined Surface when the Energy Dissipation Effects are Significant

The application of hybrid nanofluid is particularly relevant in many engineering systems where efficient heat transfer is vital. In particular, in thermal management of electronic devices, electrical chip cooling and industrial processes, the role of the combined effect of various nanoparticles is important. Therefore, this study aims to develop a machine learning-based artificial neural network (ANN) to forecast the heat transfer rate considering water-based micropolar hybrid nanofluid flow with CuO–Ag as particles flowing across a slanted surface. The free convective flow owing to the thermal buoyancy in association with the viscous, Joule and Darcy dissipation, and heat source encourages the flow phenomena. Finally, the velocity slip along with the convective heat condition affects the proposed model due to its microstructure and the consideration of thermophysical models. The flow governing equations follow a set of standard transformation rules, and further traditional shooting-based Runge–Kutta technique is employed for the solution. Graphical representations are used to analyze the characteristics of the variables causing the flow phenomenon, with validation of the results that are presented in tabular form. The novelty of the proposed study lies in the implementation of the ANN approach to predict the heat transfer rate for different factors, and the best performance results are depicted clearly. The fluid velocity is notably influenced when flowing through a permeable medium, serving as a governing factor in its behavior. Additionally, as velocity slip intensifies, a remarkable contraction in the thickness of the velocity boundary layer becomes evident. A greater Eckert number corresponds to an evident enhancement in the fluid’s overall thermal state, signifying a diminished temperature variation. The heat source parameter ([Formula: see text], Case-2) exhibits the most pronounced impact on enhancing heat transfer due to its relatively higher mean square error (MSE) value (1.0979E−12).