碳酸盐岩地层划分及类型识别方法  

Automatic segmentation and type identification of carbonate Formation

作　　者：李志虎[1] 吴培侗 张翔[1] 罗利 毛英雄 

机构地区：[1]长江大学油气资源与勘探技术教育部重点实验室,湖北武汉430100 [2]中国石油集团川庆钻探工程有限公司测井公司,重庆400021

出　　处：《断块油气田》2017年第2期169-173,共5页Fault-Block Oil & Gas Field

基　　金：国家自然科学基金项目"复杂地质背景下电成像测井层理面检测与产状快速提取方法研究"(413741148)

摘　　要：四川盆地普光地区碳酸盐岩储层具有溶蚀孔洞发育、裂缝较发育和常规测井响应特征不明显等特点,目前,利用常规测井资料进行地层划分及地层类型识别存在困难,并且现有方法仅仅只能确定地层界面,并不能有效识别地层类型。为此,文中提出了基于成像测井资料的地层划分及类型识别方法。成像电阻率数据经过处理生成平均电阻率曲线,并与自然伽马曲线进行融合,生成反映地层信息的融合曲线,进行初始地层划分;从电成像测井图像中提取熵、能量等纹理特征,与图片数据库中的电成像测井图像进行欧式距离匹配,微调初始分层界面,并确定地层类型。实例应用表明,该方法融合多种测井信息划分地层,确定地层类型,效果良好。Due to the dissolved pores development, evolution of fractures and no obvious response of conventional well logging in Puguang Area, Sichuan Basin, layers can not be effectively segmented using conventional well logging data. Since the conventional methods can only segment layers and can not recognize layer types, so an approach in this paper for automatic segmentation and layer type identification was proposed by imaging logging data. First, resistivity data is processed to generate average resistivity curve, then it is fused with GR curve to generate a fusion curve that can reflect formation information to conduct initial segmentation. Entropy and energy that extracted from resistivity logging image are matched with Euclidean distance to slightly move boundaries and confirm layer types. Results of practical examples show this method is effective and feasible.

关 键 词：破酸盐岩 成像测井 自动分层 纹理特征 地层类型识别 

分 类 号：TE122.221[石油与天然气工程—油气勘探]