Artificial-intelligence-assisted modeling of radiative viscoelastic micropolar hybrid nanofluid oscillatory flow with Cattaneo-Christov heat and mass transport

Abstract

The artificial neural network (ANN) plays a novel and innovative role in the of heat transfer by serving as a powerful predictive and data-driven modelling tool that captures complex thermal behaviors with excellent accuracy. This investigation aims to provide an optimized ANN model for oscillating flow of viscoelastic micropolar hybrid nanofluid with evaluation of heat and mass transfer. Hybrid nanofluid features characterize the suspension of aluminum alloy (AA7072-AA7075) with human blood as the base liquid. Both energy and concentration equations are modified by employing the relations of the advanced Cattaneo-Christov model. The radiated features have contributed to modifying the energy equation. Modeling of the problem leads to a nonlinear system of partial differential equations (PDEs), which are handled analytically. Based on the simulated numerical model, the ANN algorithm is implemented to predict the accuracy of the model. Understanding of machine learning algorithms has been examined through regression plots, fit functions, mean residual error, and histogram curves. Physical observations of models are revealed by evaluating the role of modeled parameters. By combining analytical simulations with ANN-driven predictive results, the analysis revealed a powerful framework for improving thermal performance, enhancing the flow stability, and enabling complex hybrid nanofluid systems. It has been observed that the developed ANN model exhibited excellent predictive capability with suitable convergence, while regression, mean square error and histogram confirmed its reliability and stable accuracy. The enhanced nanoparticles volume fraction and Brownian movement enhances the thermal impact of blood-based nanofluid. The larger micropolar fluid parameter and concentration relaxation presents a decay in the concentration diffusion.