Using multi-sensor technology

some parameters affecting the measurement of water content ratio of crude oil are measured

and two information fusion methods of multivariate regression based on nonlinear transposition of sample matrix and genetic neutral networks are presented

then being used to research the measurement of water content ratio of crude oil in high accuracy and intelligent.The assessed results show that the prediction effect of neutral networks optimized by genetic algorithm is obviously better than that of multivariate regression and neutral networks with single algorithm of LM

and take advantage of global convergence and strong generalization capability

get a good fusion effect and high measureing accuracy.All show that genetic neutral networks is an intelligent information fusion technology with extensive application prospect.