Entropy optimization of Eyring-Powell hybrid ( TiO 2 + Au )/blood nanofluid with gyrotactic microorganisms over inclined cylinder: An ANN-based study

Abstract

This work explores Eyring-Powell hybrid (TiO₂+Au)/blood nanofluid with past an inclined stretching cylinder. With the inclusion of gyrotactic microorganisms, chemical reaction, activation energy and thermal radiation. The results of this study can apply in diverse fields such as biomedical systems, thermal management and environmental applications. In this study, the similarity transformation is used to change the partial equations into ordinary differential equations. The transformed non-linear ordinary differential equations are solved by fifth order Runge-Kutta Fehlberg method with shooting technique. Furthermore, entropy generation is considered for this model, it evaluates energy losses and enhance the overall efficiency and performance of the system. In this study artificial neural network is used to confirms the reliability and accuracy of the obtained data. The Levenberg-Marquardt Backpropagation algorithm was used for training neural networks and Tan-Sigmoid function is considered for hidden layer, while the output layer uses a Purelin function. Notably, increasing suction parameter decrease the velocity profile and increment of activation energy parameter decrease the concentration profile. Increasing the thermal radiation increase the Bejan number. Moreover, thermal radiation increased the Nusselt number 57% from 0.2 to 2.0. The ANN model achieves a best validation performance for Microorganism density is 9.7118e–0.6 at 102 epoch. In the regression analysis for skin friction coefficient yields an $R$ -value approached to unity.