Application of artificial neural network for understanding multi-layer microscale transport comprising of alternate Newtonian and non-Newtonian fluids

We study the pressure driven transport of a multilayer system consisting of alternate Newtonian and non-Newtonian fluids in the narrow fluidic configuration. We consider the power law model to describe the rheology of the non-Newtonian fluid in this study. The first part of the analysis, employing the assumptions routinely conferred in the literature, constitutes for an analytical solution of the transport equation resulting in the closed-form expressions of velocity field and net throughput. We demonstrate that the rheology driven modification in the flow velocity for a given strength of the applied forcing leads to the prolific amplification in micro-pumping phenomena. In the second part, the use of artificial neural network (ANN) is implemented to present an alternative solution for the problem undertaken in this study by using the analytical results from the first part. To make the prediction using ANN, various influential transport parameters are considered and their effects on the velocity and net throughput are analyzed. It is elucidated that an appropriately trained neural network can act as a good alternative for time-consuming calculations on the multi-layered fluid flow in micro channels. The flow configuration considered in this analysis alongside the findings of our work may provide a pathway of developing novel analytical platforms for precise loading and handling of chemical samples of biological fluids in a portable fluidic environment.