Morlet Wavelet Neural Networks-Based Intelligent Approach to Analyze the Impact of Aligned Magnetic Field on a Nanofluid Thin Film Flow With Irreversibility Analysis and Chemical Reactions

This article investigates the flow and heat transfer of a nanofluid liquid film containing carbon nanotube nanoparticles over a stretching surface under the influence of an aligned magnetic field in a Darcy-Forchheimer absorbent medium. The study examines two aqueous-based nanofluid combinations: one with single-wall carbon nanotubes (SWCNTs) and the other with multi-WCNTs (MWCNTs). The choice of these nanotubes is owing to their amazing characteristics including feather-weight, remarkable thermal and electrical conductivities, and chemical and mechanical steadiness. These flows are influenced by variable nonuniform source/sink effects and thermal radiation. Furthermore, the analysis incorporates the distinct characteristics of homogeneous-heterogeneous (h-h) reactions. The novel unsupervised Morlet wavelet neural networks (MW-NNs), combined with a heuristic algorithm, are used to solve the nonlinear ordinary differential equations (ODEs). The MW function transforms the ODEs into an artificial NNs-based fitness function and then particle swarm optimization (PSO) is used for optimal fitness values. The weights of MW-NNs are optimized using PSO within the range of −10 to 10. To evaluate the convergence of this approach, fifty independent runs were performed to compute the statistical analysis for the fitness values. The results are presented through illustrations and tabulated estimates. It is witnessed that fluid velocity shows conflicting trends for the film thickness and magnetic parameters. It is also examined that the fluid temperature is enhanced for the radiation and nonuniform source/sink parameters.