西安石油大学学报（自然科学版）2024，Vol.39Issue(5)：50-58,70,10.DOI:10.3969/j.issn.1673-064X.2024.05.007
地震多属性储层厚度半定量预测
Semi-quantitative Prediction of Reservoir Thickness by 2D Seismic Multi-attributes:Taking Shan-2 Reservoir in the Southeast of Ordos Basin as an Example
摘要
Abstract
The S2 tight sandstone gas reservoir in the southeast of Ordos Basin is characterized by thin target layer,great burial depth,gentle structure,and strong reflection interfaces such as coal and limestone in the upper and lower surrounding rocks,which makes it difficult to quantitatively predict the reservoir thickness by using loess plateau 2D seismic data of low-frequency and weak-signal.A case study of semi-quantitative prediction is conducted for the Permian S23 reservoir of Yan 113-Yan 133 block in Yan'an gasfield.Using the 2D seismic data standardized before extracting attributes and the graded actual drilling reservoir thickness data,the relationship between the 2D seismic attributes and actual drilling reservoir thickness level is analyzed,and several multi-attributes semi-quantitative reservoir thickness prediction schemes are established by using random forest algorithms.The correlation coefficient between the thickness data of S23 predicted using the optimal scheme and the actual drilling results is 0.77.This study achievement deepens the geological under-standing of the distribution of the S23 reservoir,increases the geological reserves in the study area,and remarkable results are achieved in well construction.关键词
二维地震/地震多属性/随机森林算法/储层半定量预测/鄂尔多斯盆地Key words
2D seismic/seismic multiple attributes/random forest algorithm/semi-quantitative prediction of reservoir thickness/Ordos Basin分类
能源科技引用本文复制引用
俞天军,全敏,罗文琴,靳弘,万永平,唐明明,陈刚..地震多属性储层厚度半定量预测[J].西安石油大学学报（自然科学版）,2024,39(5):50-58,70,10.基金项目
国家自然科学基金"致密砂岩油藏CO2驱流固耦合效应对微观孔隙结构作用机理及表征模型构建"(52304040) （52304040）
