川东云和寨气田石炭系储集层特征及预测  

Carboniferous reservoir characteristics and prediction of Yunhezhai Gas Field in the east of Sichuan Basin

作　　者：倪根生[1] 杨斌[1] 孟昱璋[1] 周存俭[1] 

机构地区：[1]成都理工大学油气藏地质及开发工程国家重点实验室,四川成都610059

出　　处：《断块油气田》2009年第6期34-36,共3页Fault-Block Oil & Gas Field

摘　　要：云和寨气田石炭系黄龙组主要储集空间为孔隙和裂缝,属于低孔低渗型储层,而裂缝在改善储层渗透率方面发挥着重要的作用。以测井信息为基础,利用神经网络算法对该区未取心井储层的孔隙度、渗透率、含水饱和度参数及裂缝发育程度进行了预测。使用误差统计法对储层参数预测模型效果进行评价,预测效果满足本区所需储层参数计算的精度要求,证明了神经网络算法是在测井信息较少的情况下预测储层的有效手段,为气田评价井、开发井的部署、储量计算及气田开发方案的编制提供了可靠的地质依据。Pores and fractures are principal reservoir spaces of Huanglong Formation reservoir in Yunhezhai Gas Field, which belongs to the low porosity and permeability reservoir, but the fractures play an important role in improving the reservoir permeability. Based on the information of logging, the reservoir porosity, permeability, water saturation and the degree of fracture growth of uncoring well in study area have been predicted by using neural network algorithm. The effect of prediction model of reservoir parameter has been evaluated by the error statistical method, which meets the needs of calculation precision of reservoir parameters. It has proved that the neural network algorithm is an effective measure in prediction reservoir when the logging information is less. It has provided a reliable geological basis for the deployment of appraisal and development wells, calculation of reserves and compile of the development plan in the gas field.

关 键 词：碳酸盐岩 神经网络算法 储层预测 石炭系储集层 云和寨气田 

分 类 号：TE122.23[石油与天然气工程—油气勘探]