A design of machine learning algorithms for Darcy-Forchheimer flow in magnetized Carreau nanofluid flow: An advanced competent Bayesian regularization framework

With increase in world population the global energy demand is on peak, whereas traditional energy sources are not sufficient to cope the global energy needs moreover these sources cause serious pollution issues. Nanofluid took attention of researchers as it plays key role in reducing environmental pollution and support sustainable development with growth in global population, moreover applications of nano-fluids are numerous used in various fields including, pharmaceuticals, hybrid-energized mechanisms, nuclear reactors, computer processors, gas turbine blades, home freezers, cancer treatments, fuel cells and microelectronics. Here we have considered non-Newtonian fluid along with effects of nanoparticles being a specified novel importance as a consequence of their improved consumption in industrial processes, primarily in chemical engineering procedures and polymer manufacturing. In present article we have studied the heat generation/absorption and entropy generation comparison for flow of stagnation point MHD of nanofluids towards stretching sheet; furthermore, the impacts of important fluid parameter termed as thermophoresis parameter along with Brownian diffusion are under discussion for nanoparticles. Entropy generation is examined by applying second law of thermodynamics. Additionally, for fluid friction, heat as well as mass transfer, entropy generation is studied. This work is also aimed to estimate the mixed convection transmission of heat for nanofluids with inclined porous cavity with magnetic field. Role of microorganism suspended in the nanofluids are studied along with magnetic field parameter. By using transformation PDEs are converted into ODEs and graphs are plotted using bvp4c for different parameters. The result thus obtained illustrate that rate of entropy is directly related to Brinkman number, diffusion coefficient, magnetic parameter, fluid's temperature as well as concentration profile and entropy number for volume. Numerical problems are also solved by use of artificial intelligence neural networks along a special Levenberg Marquardt Backpropagation technique (BP-LMA). The effectiveness of the projected NNs of AI using BP-LMA is deliberated using a variety of performance metrics, such as Mean Squared Error, Training State of Function, Fitness State of Function, graphs for Regression Analysis and Error Histograms. The main elements effects on Microorganism profiles, Temperature profile, Microorganisms profile and Concentration profile are also thoroughly investigated. The graphs of temperature profile denoted by θ ( η ) increases as E c (The Eckert number) increases and Temperature profile declines by up surging Pr (Prandtl number). When S c (Schmidt number) grows the Concentration Profile ϕ ( η ) falls, moreover it is revealed through graphical study that by increase in L b , the resultant graph of χ ( η ) drops.