摘要
在碳酸盐岩潜山勘探中,为了解决因基础资料缺失而造成的岩相分析困难的问题,提出了基于地震属性聚类的岩相分析方法。该方法基于模糊C均值聚类算法(FCM),首先通过地震属性提取、优化和聚类分析得到地震相图;然后结合地质基础资料赋予每种地震相准确的岩相意义;最后检验结果准确性并分析岩相展布规律。选取振幅扭曲度、平均波峰振幅、最小振幅、均方根振幅、平均瞬时频率等5种属性进行FCM聚类分析,研究南堡凹陷内石灰岩、泥灰岩和白云岩3种岩相的展布特征,结果表明:石灰岩相呈基底式分布、面积约占61%,泥灰岩相多呈条带状分布、面积约占31%,白云岩相多呈点状分布、面积约占8%;在南堡1井和老堡南1井进行验证,吻合程度较高,说明该方法用于分析潜山储层岩相是可行的。
In order to tackling with the difficulties of petrographic analysis due to data absence during the exploration of carbonate buried hill,the petrographic analysis method based on seismic attributes cluster was proposed.Based on the fuzzy C-means(FCM)clustering algorithm,this method mainly contains three steps:firstly,the seismic facies map is obtained by the extraction,optimization and cluster analysis of seismic attributes;secondly,combined with geological basic data,the accurate lithofacies meanings are given to every seismic facies;lastly,the accuracies of results are tested and the distribution rules of lithofacies are analyzed.Five kinds of seismic attributes were selected to conduct the FCM cluster analysis skew in amplitude,average peak amplitude,minimum amplitude,RMS amplitude and average instantaneous frequency.Lithofacies distribution characteristics of limestone,marlite and dolomite in Nanpu sag were revealed.The results show that,the limestone covering about 61% of the total area represents the basal distribution;the marlite covers about 31% with a banding distribution;and the dolomite covers about 8% with a punctiform distribution.Actual test results of NP1 and LPN1wells indicate that this method is feasible to study the lithofacies of buried hill reservoirs.
出处
《中国矿业大学学报》
EI
CAS
CSCD
北大核心
2015年第4期688-695,共8页
Journal of China University of Mining & Technology
基金
国家油气重大专项(2011ZX05009-002)
中央高校基本科研业务费专项资金项目(14CX06068A)
中国石油大学(华东)研究生创新工程项目(CX2013003)
关键词
模糊C均值聚类
地震属性
岩相
南堡凹陷
奥陶系储层
fuzzy C-means clustering algorithm
seismic attributes
lithofacies
Nanpu sag
theOrdovician reservoir