Generation of temperature profile by cascade-forward type artificial neural network in flow of Non-Newtonian third grade fluid through two parallel plates with viscous dissipation

Abstract

Generation of temperature profile for non-Newtonian fluid flow problem by cascade-forward type artificial neural network, is reported. Solving a problem involving non-Newtonian fluid is difficult due to non linear relationship between stress and strain rate. The problem becomes more difficult under viscous dissipation considerations. Analytical solutions are useful for validation of numerical solutions. In the present problem, least square method (LSM), has been employed to solve the governing equations. The velocity and temperature profile computed by LSM is used to train a cascade –forward type artificial neural network (ANN). The trained ANN model is capable of generating temperature profile corresponding to any given velocity profile. Present work demonstrates the scope of cascade-forward type ANN model to solve these type of complicated problems of practical use. Artificial neural network (ANN) is employed in the solving of the problem with Scaled Conjugate Gradient (SCG) as training algorithm.