Application of neural networks for modelling centrifugal pumping units of booster pump stations for a two-phase gas-liquid mixture

In this paper, we investigate operational principles of booster pump stations with the aim of developing a method to calculate the influence of the cross-sectional shape of the channel on the coefficient of hydraulic friction and the effect of the curvature of the fixed channel on the coefficient of hydraulic friction and influence of rotation on the coefficient of hydraulic resistance for cylindrical rectilinear channel and border conditions. In this context, we propose the application of neural networks to build a model of centrifugal pumping units of booster pump stations for a two-phase gas-liquid mixture.