Micropolar fluid models are applied in processes where non-Newtonian fluids like polymers or colloidal suspensions move through channels with irregular and asymmetric geometries. These models are eminent in forecasting the flow behaviour. Asymmetric channels combined with micropolar fluid models help to simulate the flow of drug-laden fluids through micro-channels in drug delivery systems ensuring accurate dosage and controlled flow of drugs to targeted areas. This paper explores the application of the Levenberg-Marquardt neural network (LMNN) and quasi-linearization method (QLM) in solving an asymmetric channel flow problem incorporating micro-structured micropolar fluid. Using the micropolar fluid theory, the governing nonlinear partial differential equations (PDEs) for this flow problem are constructed subject to suitable boundary conditions that take into account the asymmetric geometry of the channel. Traditional analytical methods often fail to provide solutions for such highly nonlinear systems, necessitating the use of numerical approaches. A numerical data comparison is provided that appraises the proficiency of the code. The present methodology presents a viable solution where conventional approaches typically fall short. The angular velocity for magnetic parameter M 0 is noticed to be deteriorating before concavity (at the beginning) but after that it increases.
Hussain et al. (Thu,) studied this question.