Application of PSO-RBFNN to the Prediction of Moisture Content in Crude Oil of Wellheat Metering

Crude oil moisture content is a significant data of surface flow rate, and is also an indispensable parameter of measuring the development prospects of oilfield. During logging mining the oil field and the transportation, high precision measurement data of crude oil moisture content can optimize production parameters and improve the tar productivity. Through the related data obtained by coaxial line phase method of the moisture content meter of new online measurement device, based on influence factors of crude oil moisture content prediction, a predicting model of a particle swarm optimization of the RBF neural network for ground oil well moisture content measure is established. Simulation and experimental results show that the PSO-RBF neural network can achieve better fitting precision and prediction effect.