Designing artificial neural network of hydromagnetic bioconvection flow of micropolar nanofluid with gyrotactic microorganism past an inclined stretching sheet in a porous medium

Abstract The investigation of the thermal conductivity of nanofluids has been the central objective of an extensive number of scholars. The objective of this research is to examine the efficacy of artificial neural networks (ANNs) in predicting the heat and mass transport phenomena that occur during the bioconvection of micropolar nanofluids through a porous medium featuring a vertical stretching surface. The flow is influenced by thermal diffusion, chemical reaction, and heat generation/absorption. Buongiorno's model is successfully implemented for nanofluid phenomena, with an emphasis on Brownian motion and thermophoresis. Bioconvection pertains to gyrotactic unicellular organisms residing in aqueous environments, characterized by sporadic movement in the direction of gravity facilitated by indistinct horizontal verticilli. The discharge of micropolar fluid is symmetrical. In order to simplify the governing equations into a non‐linear differential system, self‐similar variables are introduced. These systems are then solved using neural networks based on the Levenberg–Marquardt (LM) algorithm in order to address the flow problem. The LM technique is employed in conjunction with the bvp4c numerical method to generate a dataset that encompasses various flow parameter scenarios. The Fitness summary, state transition, histogram error, and regression presentation are all further pieces of evidence that demonstrate the resilience and consistency of the LM solver. The strength of the suggested technique, where the accuracy level is attained in the ranges from 10 −8 to 10 −2 , is shown by the strong convergence between the reference solutions and the solutions generated by adding the effectiveness of a constructed solver LMB‐NN.