Stochastic supervised networks for numerical treatment of Eyring–Powell nanofluid model with Darcy Forchheimer slip flow involving bioconvection and nonlinear thermal radiation

The aim of this study is to estimate the solution of Eyring–Powell nanofluid model (EPNFM) with Darcy Forchheimer slip flow involving bioconvection and nonlinear thermal radiation by employing stupendous knacks of neural networks-based Bayesian computational intelligence (NNBCI). A dataset for the designed NNBCI is generated with Adam numerical procedure for sundry variations of EPNFM by use of several variants including slip constant, Schmidt number, mixed convection parameter, Prandtl number, and bioconvection Lewis parameter. Numerical computations of various physical parameters of interest on EPNFM are estimated with artificial intelligence-based NNBCI and compared with reference data values generated with Adam’s numerical procedure. The accuracy, efficacy, and convergence of the proposed NNBCI to successfully solve the EPNFM are endorsed through M.S.E, statistical instance distribution studies of error-histograms, and assessment of regression metric. The proposed dataset exhibits a close alignment with the reference dataset based on error analysis from level EFormula: see text to EFormula: see text authenticates the precision of the designed procedure NNBCI for solving EPNFMs. The executive and novel physical importance of parameters governing the flow, such as nanofluid velocity, temperature, and concentration profiles, are discussed. The observations imply that the presence of the slip constant, mixed convection parameter and Lewis number influences the velocity of the nanofluid. However, it is observed that temperature of the nanofluid declines for higher values of Prandl number while the concentration of nanofluid improves with increasing values of Schmidt number.