测井技术2025，Vol.49Issue(5)：748-761,14.DOI:10.16489/j.issn.1004-1338.2025.05.010
鄂北地区盒1段岩性差异化模型约束下致密砂岩储层物性分级评价研究
Study on Physical Property Classification Evaluation of Tight Sandstone Reservoir Under the Constraint of Lithology Differentiation Model of He1 Member of the Northern of the Ordos Basin
摘要
Abstract
The tight sandstone of He1 member of the northern of the Ordos basin has the characteristics of low porosity and low permeability,complex lithology and strong heterogeneity.It is difficult to achieve fine evaluation of reservoir lithology and low calculation accuracy.Therefore,physical property evaluation under the constraint of lithology differentiation model based on Bagging algorithm is proposed.Firstly,four types of reservoir lithology are divided based on rock skeleton type and structural difference,and the lithology sensitive characteristic curve is optimized by logging chart method and Pearson correlation coefficient method.Based on Bagging algorithm,the lithology differentiation model is established to realize the accurate identification of reservoir lithology.The lithology differentiation model constrains the calculation of physical parameters and combines the measured physical property data of the perforated interval to calibrate the predicted physical property data to complete the classification of Class Ⅰ,Ⅱ and Ⅲ reservoirs.The distribution law of reservoirs in the study area is clarified by sub-layers of river channels,and the fine classification and evaluation of reservoir physical properties are realized.The results show that:①The lithology differentiation recognition model based on Bagging ensemble algorithm significantly improves the lithology classification accuracy of complex reservoirs.The cross-validation results show that the lithology recognition accuracy of the model prediction group reaches 94%.On this basis,the high-precision calculation of reservoir parameters is realized by lithology difference constraint,the average accuracy of porosity calculation results and core measured data is 86.2%,and the relative error of permeability prediction results is controlled within 33%.②The porosity is greater than 9%and the permeability is greater than 0.8 mD,which is class Ⅰ reservoir.The average calculation accuracy of porosity in class Ⅰ reservoir is 89.7%,and the prediction error of permeability is less than 27%.The porosity is[6%,9%]and the permeability≥0.4 mD,the porosity>9%and the permeability is[0.4,0.8]mD,which are the type Ⅱ reservoir.The porosity is[4%,6%)and the permeability≥0.2 mD,the porosity>6%and the permeability is[0.2,0.4)mD,which are class Ⅲ reservoir.③Type Ⅰ reservoirs are mainly developed in H1-2 and H1-1 of No.3,No.4 and No.5 channels.The channel develops contiguous thick sand bodies with positive rhythm deposition.The sedimentary particles are rounded and sorted better and the particle size is coarser.The primary pores are developed and the physical properties are excellent.Type Ⅱ reservoirs are mainly developed in H1-3,H1-1 and H1-2 of No.7 and No.8 channels.Class Ⅲ reservoirs are prominent in H1-4 of No.1 and No.2 channels.It is concluded that the Bagging algorithm has a significant effect on lithology identification,and The prediction of physical properties based on the lithology differentiation model achieves a more detailed evaluation of physical properties,and clarifies the favorable exploration channels and intervals,which provides a direction for the exploration and development of natural gas in He1 member of the northern of the Ordos basin.关键词
杭锦旗地区/致密砂岩/Bagging算法/岩性差异化模型/物性分级评价/储层分布规律Key words
Hangjinqi area/tight sandstone/Bagging algorithm/lithologic differentiation model/physical property classification evaluation/reservoir distribution law分类
天文与地球科学引用本文复制引用
林行杰,王力,李可赛,刘岩,刘彦君,展硕硕,雷铠铖,陈婷,郑明明,孙佳琦..鄂北地区盒1段岩性差异化模型约束下致密砂岩储层物性分级评价研究[J].测井技术,2025,49(5):748-761,14.基金项目
国家自然科学基金项目"多尺度裂缝性储层中随钻方位电磁波测井响应机理研究"(42404144) （42404144）
