Deep Neural Network Modelling of Multi-Powered Ellis Tri-Hybrid Nanofluid Dynamics in Porous Media with Gyrotactic Microorganisms: A Study on Energy and Entropy Behavior

The main objective of the current endeavor is to monitor hypothetical events utilizing an Ellis tri-hybrid fluid submerged in gyrotactic swimming microorganisms and exposed to a rotating disk with entropy production using deep neural network-based Adam optimization. In order to conceptualize an R-K 4th order with firing method over a porous material, EMHD and non-linear thermal radiation are considered. Further, the physical attributes of a fluid, including skin friction coefficients, Nusselt number, and motile microorganisms, are evaluated using a DNN-based Adam optimisation model. Self-similarity is a transformation that helps to reduce a highly non-linear PDE set of equations into an ODE. The model’s study results are largely consistent with previous research, with a few notable deviations. Graphically displayed the results for various distributions, drawing inspiration from active components. The azimuthal and radial velocity imaging improves as the parameter of the Ellis fluid increases. Beneficial effects and thermal radiation boost the temperature, much as an increase in the bio-convection Lewis number affects fluid motility. As the Bejan number of the electric field increases, so does the entropy production of the temperature ratio parameter. In addition, DNN delivered the most accurate values for the physical attributes of a fluid. Nanofluid, which are composed of water and nanoparticles such as copper, titanium and carbon nanotubes, improve cooling in solar heaters, computers, motors, and machining. They are also used in nuclear reactor cooling, electric battery production, water purification, and cancer therapy.