Numerical and artificial intelligence analysis of squeezing flow between parallel disks under complex physical fields.

This study presents the time-dependent squeezing flow between parallel discs under the influence of various physical processes such as magnetohydrodynamics, micropolarity, thermophysics, homogeneous-heterogeneous chemical reactions, and bioconvection. The slip and permeability effects are considered at the boundaries. The nonlinear system is simplified through scaling transformations and subsequently solved numerically. Besides, an artificial neural network (ANN) model is developed to predict the values of skin-friction coefficients, heat transfer rates, and mass transfer rates at both discs. The performance of the ANN model is validated against the computational results, demonstrating robustness with correlation coefficients being almost close to 1. The ANN model performs more robustly for instance, skin-friction coefficient model revealed correlation (R) equaling 0.99775 and 0.99537 for the test data of the lower and upper discs, respectively. The values for heat and mass transfer observe as 0.99999, 0.99938, and 0.99839, 0.99167 for the test data of lower and upper discs. Parametric and sensitivity analyses produce results that are comparable to the numerical findings. Graphical and tabular representations are provided for various dimensionless parameters. Results indicate that the developed model is reliable and capable of accurately predicting multiple physical processes. [ABSTRACT FROM AUTHOR]