Artificial neural networks analysis of an unsteady fluid flow through a horizontal oscillating flat plate in a porous slab

The current paper investigates an artificial neural network (ANN) analysis of an unsteady thermo-viscous fluid with incompressible motion in a porous slab in a region of semi-infinite through an horizontal oscillating flat plate with an impermeable bottom. Suitable conditions at the boundary were applied to get the solutions of temperature and velocity distributions. The analytical results were obtained for the heat transfer coefficient, shear stress, and external forces. The feed-forward neural network algorithms are also utilized to compute and analyze the solution of governed flow of the fluid. A neural network with multi-layer perceptron is applied for trial functions, and the adaptive moment estimation algorithm is utilized to estimate the suitable parameters with adjustable capability. The numerical computations of both exact and ANN solutions have been presented in tabular form and visually depicted for various physical factor values. The convergence analysis of the problem has been studied with respect to the velocity and temperature on the plate boundary with the various values of data points and hidden neurons. The effectivity of the solution increases with the growth of neurons and data points within the neural networks. The R-squared parameters with high values were achieved to estimate the velocity (0.999 86) and temperature (0.999 95) of the fluid. Moreover, because it takes less time and processing power to solve the problem, the current ANN model may be used to more complicated models.