A machine learning framework for quantifying microbial dynamics in hybrid (ZnO-CoFe₂O₄) flows

This study introduces a machine learning framework to quantify the motile microbes within the hybrid structured flows under the prominent impact of chemical reaction. The main objective of this research is to examine the chemical, thermal and biological characteristics of the moving microorganisms in the flow of pure and hybrid fluids. The nanofluids ZnO-CoFe₂O₄/C₃H₈O₂ and ZnO/C₃H₈O₂ are prepared using propylene glycol as the host fluid. The preeminent influences of microbial motility and nanoparticle dispersion on biological interactions and the effectiveness of heat transfer in nanostructured fluids are explored. A machine learning method developed in MATLAB is used to forecast and optimize the thermal performance of both mono and hybrid nanofluids. The ML and numerical results are found to be in good agreement under several conditions. Quantitative analysis shows that the inclusion of Forchheimer inertia decreases the heat transfer rate by approximately 15%. The hybrid nanofluid (ZnO–CoFe₂O₄/C₃H₈O₂) improves thermal transport by nearly 12% compared to the single nanofluid. In addition, increasing the bioconvection Peclet number results in a 30% enhancement in microorganism flux. Whereas stronger chemical reaction reduces mass transfer by about 20%. The predictions from the ANN-based NF-tool closely match the numerical results with absolute errors remaining below 10 −4 . It assures the accuracy of the proposed hybrid numerical-machine learning framework. The results highlight the potential of ZnO and ZnO-CoFe₂O₄ nanofluids in advanced thermal management applications such as industrial processes, environmental systems, and bioengineering.