Life Prediction of Slotted Screen Based on Back-Propagation Neural Network

Abstract Slotted screen is widely used in sand control, but screen erosion life seriously affects its performance. In this paper, the computational fluid dynamics (CFD) method was used to calculate the erosion rate of the single slot. The wall erosion of the slot has been analyzed on the field condition of production pressure difference, dynamic viscosity, particle size distribution and sand content. Based on BP neural network, a service life prediction model of slotted screen erosion has been established and was used to calculate the screen service life by the field conditions. The results showed that: (1) By the influence of the screen erosion, the kinematic viscosity and the particle diameter each has a turning point. But they have the different affecting trend as the data growing. (2) It’s a good method by using Back-Propagation (BP) neural network to get the slotted screen life prediction model. The tolerance between the predicted data based on the Predictive model and the calculated data is less than 15%. The service life of the slotted screen is calculated according to the field conditions of the Bohai Bay, the resulting life change situation is consistent with the actual situation of the slotted screen. The established life prediction model of slotted screen has certain reference significance.