摘要
多点地质统计学方法(MPG)借助"训练图像",考虑了空间多点间的信息,能更好地反映复杂目标体的几何结构。该方法获得的岩相模拟结果忠实于井数据,垂向分辨率较高;以概率形式融合地震信息,横向分辨率较高。为了进一步提高模拟结果的横向分辨率,研究提出采用基于Tau模型的多属性概率融合算法进行多属性处理。首先介绍了多点地质统计学的基本原理,然后提出了新的多属性概率融合算法,最后针对民丰洼陷地区,系统开展了多属性约束的多点地质统计岩相预测方法研究,并取得了有效的成果。
Abstract. Multipoint geostatistics （MPG） are used to make lithofacies prediction for complex geology, which considers the correlations between more than two locations in space by means of Training image. The method has very high-resolution as the logging data and seismic data serves as the hard and soft data in simulation. In order to improve the lateral resolution luther, we propose multiattribute fusion based on Tau model. In the paper, we first introduce the theory of multipoint geostatistics, then propose a new multiattribute fusion based on Tau model, and finally apply the methods to Minfeng Sag for lithofacies pre- diction, which has achieved some good results.
出处
《地质科技情报》
CSCD
北大核心
2017年第6期273-278,共6页
Geological Science and Technology Information
关键词
多属性
薄互层
多点地质统计
概率融合
民丰洼陷
multi-attribute
thin interbeded
multipoint geostatistics
probability fusion
Minfeng Sag
