摘要
致密砂岩储层孔隙度低、渗透率低、非均质性强,气层所对应的测井响应特征较为复杂,气层识别和评价难度较大、多解性突出。传统上,利用常规测井曲线进行含气性评价多是定性评价,在利用智能识别法评价含气性时,也是利用分类模型进行定性评价;而利用常规测井资料定量评价含气性比较困难。本文首先以岩心、地质、试气资料和常规测井曲线为基础,利用交会图法进行致密砂岩含气特征分析,建立含气性定性评价指标;然后,利用广义回归神经网络(GRNN)预测含气量和含水量,构造含气性和含水性指示曲线,定量评价致密砂岩的含气性;最后,定性评价和定量评价综合使用,以评价致密砂岩含气性,并在苏里格地区盒8段进行应用,取得了较好的应用效果。
Tight sandstone reservoirs always show the characteristics,such as, low porosity, low permeability,and strong heterogeneity. The logging response characteristics corresponding to gas-bearing reservoir is very complex,so that the identification and evaluation of gas are difficult and it always shows multiple solution. Conventional well log is used for qualitative evaluating the gas-bearing characteristics. However,the classification model in intelligent recognition method still belongs to qualitative evaluation,which makes the traditional well log difficult to quantitatively evaluate gas. We use the core,geology,gas testing and conventional logging data to analyze the gas characteristics by cross-plot method and build two indexes,which can be used for qualitatively evaluating the gas containing. The workflow includes the generalized regression neural network（ GRNN） that reconstructs gas and water indication curve to quantitatively evaluate the gas containing. Finally,we used the index method and GRNN curve reconstruction method to evaluate the tight sandstone gas in the Sulige area. The results show the good application effects.
出处
《吉林大学学报（地球科学版）》
EI
CAS
CSCD
北大核心
2016年第3期930-937,共8页
Journal of Jilin University：Earth Science Edition
基金
国家自然科学基金项目(41174096)
国家"十二五"重大科技专项(2011ZX05040-002)~~
关键词
致密砂岩
含气性定量评价
曲线重构
GRNN
指标法
苏里格地区
tight sandstone
gas-bearing quantitativeevaluation
curve reconstruction
GRNN
index method
Sulige area
