Artificial neural network based computational analysis of hybrid nanofluid flow containing gold and zinc nanoparticles

The gold nanoparticles are beneficial in cancer treatment using photothermal therapy, which uses near-infrared light absorption and conversion into heat to kill cancer cells. When coupled with other metals, like zinc nanoparticles, the anticancer action of gold nanoparticles is enhanced. Moreover, saddle points guide flow away from healthy tissues, increasing therapeutic accuracy and reducing side effects, whereas nodal points assist in concentrating medications or particles at specific areas, such as tumors. Therefore, this study is carried out to examine how thermal radiation affects the gold-zinc/blood flow at the nodal and saddle stagnation points. Tumoral tissue-based hybrid nanofluid flows through a rounded cylinder with a sinusoidal radius modification. The convective heat transfer phenomenon is analyzed based on the Tiwari-Das nanofluid model. Tumoral tissue blood is used as base fluid with gold and zinc nanoparticles. The formulated system of partial differential equations with subjective boundary conditions is transmuted into the dimensionless structure by using the similarity variables. The reduced differential equations are resolved numerically by utilizing bvp4c in MATLAB software with a shooting algorithm. Moreover, the present work presents a unique implementation of an intelligent numerical computing solution based on a multi-layer perceptron, feed-forward back-propagation artificial neural network (ANN) using the Levenberg-Marquardt algorithm (LMA). Higher values of the nodal/saddle indicative parameters increase the velocity field. The proposed ANN model converges quickly, requires no linearization, and reduces processing costs.