Levenberg-marquardt design for analysis of maxwell fluid flow on ternary hybrid nanoparticles passing over a riga plate under convective boundary conditions

Abstract

This study provides an examination of the effects of the Maxwell fluid on ternary hybrid nanoparticles passing through a Riga plate under convective boundary conditions using neural network backpropagation. Water is combined with the ternary nanoparticles Al2O3,GrapheneandCNT as the base fluid. Heat sources and sinks as well as thermal radiation are taken into account in the heat transfer study. Partial differential equations can be transformed into a form without any dimensions by applying a similarity transformation. The Finite element method is then employed to resolve this altered equation. The problem's highly nonlinear equations are numerically resolved by utilizing techniques from the Artificial Neural Network method. The Artificial Neural Network method solution technique is used to solve these ordinary differential equations and create the Neural Network Backpropagation reference dataset. The plots illustrating the solution and error analysis for varying different parameters are analyzed in MATLAB using this reference dataset. Regression analysis, error histogram, and mean squared error data are used to assess neural network backpropagation performance. The techniques for testing, validating, and training are applied to the model solution. The impacts of fluid flow and thermal field were examined through important factors. It is found that velocity field shows an opposite pattern for Maxwell fluid variable, it is an increasing function of the modified Hartmann number and the temperature distribution is more influenced by the Biot number and thermal radiation.

Keywords

• Non-newtonian fluid
• Nanofluid
• Ternary hybrid nanoparticles
• Convective boundary condition
• Artificial neural network computation
• Riga Plate: Magnetic field