Artificial neuron based repercussions of thermal prediction of steady nanofluid with involvement of antimicrobial agent in time dependent magnetized environment: blending of two models and geometries
摘要
The analysis of motile microorganisms using two distinct non-Newtonian models is crucial. Thus, the motion and interaction of microorganisms can be investigated more deeply. Two distinct models can capture a broader spectrum of fluid behaviors and predict the microorganisms' dynamics more accurately. This research depicts to harness the capabilities of neural network architecture to predict thermal properties of motile microorganisms with two distinct non-Newtonian mathematical models over the static and moving wedge shape geometries. In this attempt, neural intelligent architecture is used for thermal prediction and streamline analysis. Velocity of nanofluid is scrutinized through variable magnetic fields. Two unique models, cross and Casson, are combined in the study to capture multiple aspects of flow with two geometries. Furthermore, many physical parameters are investigated for velocity, temperature and microorganism profiles. Cross fluid models are very important because they can investigate the fluid properties at very high and low shear rate. With the help of physical assumptions, the governing equations based on assumed models generated the set of Partial Differential Equations (PDEs) and then these are changed into Ordinary Differential Equations (ODEs) using similarity variables. Solution is obtained through bvp4c execution and further, solution data set is passed under the Bayesian Regularization Neural Networks (BRNNs). The velocity profile of the Casson–Cross nanofluid decreases as