摘要
在复杂岩性储层中 ,储层四性关系比较复杂 ,表现为非线性 ,用传统的方法已经难以解决这类问题 ,为此引入了目前比较流行的人工神经网络技术 .在前人基础上 ,以SN油田 9井区为例进行储层研究工作 ,该区非均质性比较强 ,经对该区 3 64口井常规测井资料进行储层参数重新解释 ,并做平面展布 ,与实际资料吻合较好 .由此表明 ,神经网络技术在解决非线性问题上表现出了较大的优越性 。
The relation between parameters of reservoir and attributes of logging data is complex in reservoirs of complicated lithology. These non linear problems can't be solved by traditional methods, so popular manual neural network technique is adopted. The author started this study on the ninth area of SN oil field on the base of the others. The geological condition of this area is not homogeneous and complex. Data of 314 routine logging well is interpreted by this way, and the result accords with actual data well. We can see that neural network is in the ascendant in solving non-linear problems, which need us to do father work.
出处
《地球物理学进展》
CSCD
2003年第1期44-48,共5页
Progress in Geophysics
基金
中国科学院知识创新重大项目 (KZCX1 SW 1 8)资助
关键词
储层
神经网络
测井解释
reservoir, neural network, logging interpretation
