STUDY ON PREDICTION OF SATURATES OF VACUUM GAS OIL (VGO) BY USING ARTIFICIAL NEURAL NETWORKS

摘要：

Prediction on saturates of vacuum gas oil with 5 properties, average boiling point, density at 20℃, carbon residue, molecular weight and refractive index at 70℃, using neural network was developed. Comparing the calculating data with the experimental data, the average relative deviations of VGO saturates were 4.97％. Six testing samples which have been trained were predicted by using this model and the average relative deviation of the prediction were 3.91％. The accuracy of the prediction model was promising and it is available in pre liminary prediction of basic physical properties of crude oil.

关键词： VGO saturates base physical property artificial neural network prediction

作者: Renjin Sun Shouchun Wang Suoqi Zhao

作者单位: China University of Petroleum, Beijing, 102249, China

会议类型: 国际会议

会议名称: The Ninth International Conference on Industrial Management(第九届工业管理国际会议 ICIM2008)

会议地点: 日本大阪

会议语种:英文

页码: 805-811

在线出版日期: 2008-09-16（万方平台首次上网日期，不代表论文的发表时间）