摘要
针对碳酸盐岩薄储层气藏预测受制于地震分辨率,用常规反演方法往往很难解决薄储层气藏预测问题,通过参数加权综合法研究,从地震信息中提取薄储层5大类共计35种地震波场特征参数,通过Karhune-Loeve变换压缩降维,采用广义线性分类判别及加权综合处理,可获得能反映薄储层气藏特征的综合参数,较好地解决薄储层气藏预测的难题。在实际推广应用中也见到了良好的效果,如四川同福场构造嘉陵江组嘉二1气藏为一典型碳酸盐岩薄储层气藏,储层纵向上厚度较薄,一般有效储层厚度多为1m～3m,曾采用地震速度反演方法做过多轮储层预测,但因储层测井差异性特征不明显以及储层太薄等因素,预测效果均不理想,而采用地震参数综合法则较好地解决了本区薄储层气藏的预测难题,预测符合率达到95.5%。
It is very difficult to predict thin and interbedded is used to attract gas reservoirs of carbonate rock with traditional inversion method.In this paper,parameter weighted synthesis is used to attract up to 35 seismic wave field characteristic parameters from seismic information of thin and interbedded reservoir and to compress and lower dimension through Karhune-Loeve commutation.The generalized linear sorter and weighted synthesis are employed obtain synthesis parameter,to reflect the character of thin and interbedded reservoir gas reservoirs,and to solve the difficult problem of predicting thin and interbedded reservoir. This method was applied to the typical thin and interbedded gas reservoirs of carbonate rock of Tongfuchang structure of Sichuan,which has a thickness of 1 m to 3 m.This method proved to be satisfactory with the coincidence rate of prediction reaching up to 95.5%.
出处
《矿物岩石》
CAS
CSCD
北大核心
2005年第1期117-120,共4页
Mineralogy and Petrology
