Numerical investigation of constant and variable fluid properties for 3-D stagnation point flow along a moving surface: An ANN approach

Abstract

Three-dimensional stagnation point flow along a vertical sheet has immense theoretical and practical applications ranging from heat exchangers to aerodynamics. The thesis aims to discuss two different interesting types of flow problems namely (i) Three dimensional stagnation point flow for variable physical properties with ANN validation and (ii) Three-dimensional stagnation point flow of nanofluids using Buongiorno model. In the first problem, a study was conducted to examine the impact of variable fluid properties in mixed convection three-dimensional flow of viscous fluid along a verti cal sheet with heat transfer. The mathematical model is developed by considering temperature-dependent variations in viscosity and thermal conductivity. By using ap propriate transformation, the governing equations construct system of equations that permit self-similar solutions. The flow must satisfy ordinary differential equations whose solution relies on different parameters such as mixed convection parameter λ, variable viscosity θr and c1 show the 3-dimensional motion of flow. A novel approach using Artificial Neural Networks (ANN) to accurately predict heat transfer parameters. The ANNmodelistrained using a comprehensive dataset obtained from numerical sim ulations and experimental measurements. The inputs to the ANN include relevant flow parameters such as Reynolds number, Prandtl number and geometrical characteristics while the outputs are the corresponding skin friction and Nusselt number. The perfor mance of the ANN model is evaluated using both training and validation datasets. An analysis is conducted to identify the key factors influencing the prediction accuracy of the ANN model. The graphical representation of emergent parameters has been ex plored, along with a corresponding discussion. The comparison is being made between the effects of constant and variable fluid properties. The skin friction and Nusselt number are calculated along the axis. In the second problem a Buongiorno model is used in which the combined effects of Brownion motion and thermophoretic diffusion of nanoparticles are described. The obtained PDEs are transformed into ODEs us ing similarity transformation along with boundary conditions. Further, the graphical interpretation of different parameters c1, Nb, Nt and Sc have been discussed. Quanti tative analysis of the flow field and heat transfer characteristics in both problems are conducted using graphs and numerical values obtained through Bvp4c MATLAB.