Modeling of the discharge coefficient of differential pressure flowmeters: approximation by radial basis function networks

The article considers a way to model the discharge coefficient of differential pressure flowmeters. The relevance of modeling this coefficient using machine learning methods, specifically neural networks, is noted. It is proposed to use radial basis function networks to approximate the discharge coefficient of standard orifice plates. The structure of a radial basis function network was developed to calculate the discharge coefficients of an orifice plate, with an angular pressure tapping method. The article estimates the error in approximating the discharge coefficient by radial basis function networks and gives recommendations on the construction of radial basis function networks for solving problems associated with modeling the characteristics of differential pressure flowmeters. The main advantages and disadvantages of using such networks for modeling the discharge coefficients of orifice plates in differential pressure flowmeters are considered. The study confirmed the effectiveness of using radial basis function networks to approximate the discharge coefficient. The obtained results can be used to improve the accuracy of gas and liquid flow measurements using differential pressure flowmeters.