Exploring bioconvection dynamics within an inclined porous annulus: Integration of CFD and AI on the synergistic effects of hybrid nanofluids, oxytactic microorganisms, and magnetic field

The current research integrates artificial intelligence (AI) with computational fluid dynamics (CFD) to explore the magneto-bioconvective phenomenon of hybrid nanofluid within an inclined porous annulus containing oxytactic microorganisms. Initially, we analyze bioconvection dynamics and energy distribution by simulating the nonlinear governing equations. Owing to the involvement of multiple variable parameters, simulation technique demands significant resources and time to determine the optimal parametric values. To surmount this challenge, CFD-ANN-GA method is proposed to identify the optimum parameter values. Leveraging 212 CFD dataset, we developed an artificial neural network (ANN) model, tested with new dataset, and achieved an accuracy of 99.03 % and 94.82 % respectively, for average Nusselt and Sherwood numbers. This highly accurate ANN model's predictions unveiled significant insights that remained elusive from the CFD data. The CFD data suggested that the parameter set: Ra=10⁶,R_b=10,Ha=0,Da=10⁻¹,Φ=45^°,Pe=1,Le=1, experience maximum heat dissipation. However, the genetic algorithm (GA) recommended a different parameter set: Ra=10⁶,R_b=10,Ha=0,Da=10^−1.5644,Φ=40.67^°,Pe=0.100388,Le=1, providing maximum thermal transport. Subsequently, the GA recommended parameter set has been tested through simulation technique and found to exhibit comparatively greater thermal transport. Furthermore, we identified the hierarchy of parameters influence on heat and oxygen mass transport within the enclosure by conducting sensitivity analysis.