Abstract
The aim of this article is to explore incompressible and steady hybrid nanofluid flow in Darcy-Forchheimer porous medium (ISHN-DFPM) with entropy production along a stretchable surface via the strength of stochastic numerical computing by incorporating the applicability and effectiveness of artificial neural networks with Levenberg-Marquardt backpropagation (ANNs-LMB). The reference dataset is constructed in MATHEMATICA using the ND Solve command. Moreover, the generated dataset is further exploited after generating in the Mathematica software to export in the MATLAB environment under the functionality and performance of the Levenberg-Marquardt backpropagation technique. The results verification is presented by comparing the model with already perfumed simulations. The thermal onsets of the hybrid nanofluid model are characterized by using the molybdenum disulfide and silicon dioxide nanoparticles with water base fluid. The error analysis for the implemented algorithm is also presented. The numerical outcomes are listed for Nusselt number and wall shear forces against and nanoparticles. It is observed that the heat transfer phenomenon enhanced with the increasing Brinkman number. Moreover, the Bejan number and entropy generation phenomenon increased with the radiation parameters. The dataset is prepared in terms of training, testing, and validation processes to enterprise the suggested intelligent solver ANNs-LMB which is further authenticated through convergence graphs of the fitness-dependent mean square error, histogram error, and regression measure which also ensure high accuracy and predictive strength of the solver.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call