摘要
在关键井的测井数据、取芯资料及地质信息分析的基础上,研究曲流河沉积相各沉积微相的特征及在测井曲线上的响应特征,建立曲流河沉积相各种沉积微相的测井相模式及特征参数样本,利用BP(Back Propagation)神经网络技术反馈学习,得出一套适合研究区域沉积微相的判别系数,并对其它实际测井资料进行沉积微相自动识别,得出结果与地质专家解释结果吻合率达到85%以上,效果显著,实现了根据测井资料用计算机自动识别井剖面地层的沉积微相。
Based on key well logging data, core data and geological information,by researching the characteristics of snaking stream deposition microfacies and the response characteristic in logging data, electrofacies models and characteristic parameters sample are estab-lished,and then the discrimination coefficient fitted the region can be gained by using BP neural network technology. Consequently, by using the discrimination coefficient auto - identified actual log data and gain the sedimentary microfacies. Compared with the results explained by geological experts, the tallying rate of the outcome is more than 85%. It achieves the result that the sedimentary microfacies of well section can be auto-identified by computer based on logging data.
出处
《工程地球物理学报》
2009年第3期332-337,共6页
Chinese Journal of Engineering Geophysics
关键词
沉积微相
BP神经网络
判别系数
sedimentary
microfacies
BP neural network
discrimination coefficient
作者简介
卢松（1985-），男，中国地质大学（武汉）地球物理与空间技术学院在读硕士，研究方向为地球物理测井及相关解释。E-mail：slu2820213@gmail.com
潘和平（1953-），男，教授，博士生导师，主要从事地球物理测井的教学与研究。E-mail：pangpingge@163.com
