Application of FOA-GRNN to Prediction of Moisture Content in Crude Oil of Wellheat Metering

Accurate prediction in crude oil moisture content and the high accuracy data of crude oil moisture content has vital importance in oil well measurement.The moisture content prediction can be affected by many factors,and there is complicated nonlinear relationship between the prediction and factors,so the traditional forecast methods can not meet the requirements of prediction accuracy.In order to improve the prediction precision,a crude oil moisture content measure method named fruit fly optimization algorithm of the generalized regression neural network forecast method was put forward,which was used for generalized regression neural network parameters optimization.By determing several parameters which can affect crude oil moisture content using the coaxial line phase indicator measurement system,we established fruit fly optimization algorithm of the generalized regression neural network predicting model of crude oil moisture content.Simulation and experimental results show that: relative to the widely used Back Propagation Neural Network(BPNN) prediction model,the ffruit fly optimization algorithm of the generalized regression neural network,has higher prediction accuracy.It is a practical and effective prediction method of crude oil moisture content.