FINITE DIFFERENCE AND ARTIFICIAL NEURAL NETWORK APPROACH TO FRACTIONAL MHD CASSON FLUID FLOW OSCILLATION

Fractional derivatives have demonstrated a wide range of solutions beneficial for engineering, medical, and manufacturing sciences. Research on the use of fractional derivatives in fluid flow problems is still emerging, especially in analytical studies. In this paper, a mathematical model is developed to analyze the behavior of Casson fluid in heat and mass transfer processes within a magnetohydrodynamic (MHD) environment over an unsteady oscillating surface. Additionally, the numerical computation of fractional MHD Casson fluid flow for parallel surfaces subject to oscillating boundary conditions is discussed. The model utilizes a non-singular fractional derivative to solve fractional-order partial differential equations. By applying an appropriate transformation, these equations are converted into a dimensionless form. Subsequently, the finite difference method is employed to analyze the system’s behavior. A comprehensive parametric study is conducted to explore the effects of different physical parameters on the flow and thermal fields, and numerical simulations identify the fractional derivative operator that yields the most accurate and stable solutions. Furthermore, the model is solved using an ANN approach, and multiple cases are discussed for varying values of the fractional order parameter [Formula: see text]. This ANN-based analysis confirms the effectiveness of the fractional model in capturing the complex dynamics of MHD Casson fluid flow under oscillating boundary conditions.