基于层次聚类元对象表征的岩性识别方法  

Lithology identification method based on hierarchical clustering and meta-object representation

作　　者：韩建[1] 万川 曹志民[1] 郭颖[1] 段朝辉 李林 Han Jian;Wan Chuan;Cao Zhimin;Guo Ying;Duan Chaohui;Li Lin(School of Physics and Electronic Engineering,Northeast Petroleum University,Daqing 163318,China)

机构地区：[1]东北石油大学物理与电子工程学院,大庆163318

出　　处：《电子测量技术》2021年第4期104-109,共6页Electronic Measurement Technology

基　　金：国家自然科学基金项目(51574087);东北石油大学研究生创新科研项目(JYCXCX092018)资助。

摘　　要：由于测井数据标签的连续性特点,数据样本间具有较强的上下文关联性。针对现有岩性识别方法所构造基本识别单元无法充分利用测井信号曲线连续性提供的上下文信息的问题,提出了一种基于层次聚类元对象表征的岩性识别方法。该方法依据基于区域生长的分层聚类方法,综合利用多条常规测井曲线进行目标储层自动分层,然后从统计特征和形态特征双视角度实现元对象的完备表征,在提取特征后形成的丰富的特征空间上进行岩性识别。通过大庆油田实际测井数据岩性识别的对比实验,采用所提方法的实验组各类岩性识别性能均得到显著提高。Due to the continuous characteristics of logging data tags, data samples have strong contextual relevance. Aiming at the problem that the basic identification unit constructed by the existing lithology identification methods cannot make full use of the context information provided by the continuity of logging curves, a lithology identification method based on hierarchical clustering meta-object representation is proposed. This method is based on the hierarchical clustering method based on regional growth, comprehensively using multiple conventional logging curves to automatically stratify the target reservoir, and then realize the complete characterization of the meta-object from the perspective of statistical and morphological features. After the features are extracted lithology identification is performed on the rich feature space formed. Through the comparative experiment of lithology identification with actual logging data in Daqing Oilfield, the performance of various lithology identification of the experimental group using the proposed method has been significantly improved.

关 键 词：测井曲线 层次聚类 自动分层 岩性识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]