神经网络法在油藏埋存CO_2效果预测中的应用  

Application of Artificial Neural Network in Forecast of Carbon Dioxide Storage in Reservoir

作　　者：张延旭 姜晶 王涛[3] 王凯 ZHANG Yanxu;JIANG Jing;WANG Tao;WANG Kai(Pengbo Operating Company of CNOOC （China） Co. Ltd. , Tianjin 300459;CNOOC （China）Co. Ltd. Tianjin Company, Tianjin 300459;China Oilfield Services Limited, Tianjin 300450;Engineering and Technology Company of CNOOC Energy Development Co. Ltd. , Tianjin 300452)

机构地区：[1]中海石油(中国)有限公司蓬勃作业公司,天津300459 [2]中海石油(中国)有限公司天津分公司,天津300459 [3]中海油田服务股份有限公司,天津300450 [4]中海油能源发展股份有限公司工程技术分公司,天津300452

出　　处：《精细石油化工进展》2018年第2期29-32,共4页Advances in Fine Petrochemicals

摘　　要：采用BP神经网络法对油藏埋存CO_2效果进行评价预测,且引入影响埋存效果的5个无因次变量。结果表明,神经网络方法具有更好的自适应性,能较好的反映影响各影响因素与埋存系数的内在联系,且预测精度较高。因此,认为应用BP神经网络方法评价油藏埋存CO_2能力是可行和有效地。The effect of CO_2 storage in reservoir was evaluated and forecast with BP artificial neural network technique,and five dimensionless variables that affect storage effect were introduced. The results of study have shown that the artificial neural network technique has good adaptability and is capable of reflecting the inner link between the affecting factors and the storage coefficient,and its forecast precision is high. For this reason it is believed that it is feasible and effective to evaluate the CO_2 storage capacity of reservoir with BP artificial neural network technique.

关 键 词：神经网络 CO2 埋存 评价 油藏 预测 

分 类 号：TE34[石油与天然气工程—油气田开发工程] TP183[自动化与计算机技术—控制理论与控制工程]