Neural network identification of liquid flow of pump stations with asynchronous electric drive for oil refining productions.

The article discusses a method for determining the flow rate of a liquid pumping station. The liquid transfer station is a centrifugal pumping unit with an asynchronous electric drive. The properties and parameters of the pumped liquid are close to the properties and parameters of water. Determining the flow rate of centrifugal plants is an important task for efficient operation, both in terms of reliability and energy efficiency. In this work, a pre-trained neural network was used to determine the flow rate of the pumped liquid. The voltage and current signals of the stator of an asynchronous motor were used as parameters of the input layer. The results of numerical experiments are also presented, which were obtained using the developed simulation model of the installation. The article provides estimates of the dynamic and static error of the identifier of the volumetric flow rate of a centrifugal pump with an asynchronous electric drive. [ABSTRACT FROM AUTHOR]