基于岩石物理相的深层页岩气地质—工程甜点参数测井评价方法——以四川盆地LZ区块五峰组—龙马溪组为例

海相深层页岩气(埋深大于3500 m)储层受沉积、成岩、构造及生物等多种作用影响，甜点成因机制十分复杂，通常的沉积微相、岩相划分难以精细刻画其强非均质性，进一步加大了地质—工程甜点参数测井精细评价的难度。为此，以川南LZ区块页岩气储层为例，在明确海相页岩岩石物理相内涵、影响因素及核心指标的基础上，利用岩石薄片、全岩衍射以及TOC等资料，结合测井响应特征，划分出三类岩石物理相(Ⅰ、Ⅱ、Ⅲ),进一步分为6个亚类(Ⅰ₁、Ⅰ₂、Ⅰ₃、Ⅱ₁、Ⅱ₂、Ⅲ),明确了不同类型的特征及优势相，并采用随机森林算法建立了连续测井剖面岩石物理相识别方法和基于岩石物理相分类的地质—工程甜点参数测井精细计算模型。结果表明：(1)发育在五峰组顶部和龙一₁¹—龙一₁³小层内的Ⅰ₁和龙一₁⁴小层内的Ⅱ₁为有利相，通常具有TOC高、孔隙度大、...

Logging evaluation method of geological-engineering sweet spot parameters for deep shale gas based on petrophysical facies: A case study of the Wufeng-Longmaxi Formation in LZ block of Sichuan Basin

Marine deep shale gas (with a burial depth of greater than 3500 m) reservoirs are affected by deposition, diagenesis, tectonism, biological processes, and other factors, and the formation mechanism of their sweet spots is very complicated. As a result, conventional divisions of sedimentary microfacies and lithofacies are difficult to carefully describe their strong heterogeneity,which further increases the difficulty in the fine logging evaluation of geological-engineering sweet spot parameters.Therefore, this paper takes the shale gas reservoir in LZ block in the south of Sichuan Basin as an example and analyzes the connotation, factors,and core indicators of petrophysical facies of marine shales.Then,with rock thin sections, whole-rock diffraction,total organic carbon (TOC) data,and logging response characteristics,the paper obtains three types of petrophysical facies(Ⅰ,Ⅱ,and Ⅲ) and further divides them into six subcategories (Ⅰ₁,Ⅰ₂,Ⅰ₃,Ⅱ₁,Ⅱ₂,and Ⅲ),so as to explore their characteristics and dominant facies. In addition, the random forest algorithm is used to develop a method for identifying petrophysical facies in continuous logging profiles and establish a fine logging calculation model of geological-engineering sweet spot parameters based on different types of petrophysical facies.The results indicate that:1Ⅰ₁ deve-loped at the top of Wufeng Formation and the first and third sub-members of the first member of Longmaxi Formation, namely,l₁¹-l₁³,and Ⅱ₁ developed at the fourth sub-member of the first member of Longmaxi Formation ①₁⁴ are favorable facies, and they usually feature high TOC, large porosity, rich gas content, and strong brittleness.②The logging identification results of petrophysical facies based on the random forest classification algorithm are better than that of conventional logging crossplot identification methods, and the accuracy rate is above 90%.③The calculation accuracy of geological-engineering sweet spot parameters based on a random forest regression model with classified petrophysical facies is high, and the correlation coefficient of calculated and measured parameters all exceeds 0.9. The proposed method effectively solves the difficulty in obtaining geological-engineering sweet spot parameters of marine deep shale gas reservoirs, realizes the accurate calculation of well profiles of sweet spots, and lays a foundation for selecting favorable layers and optimal targets of horizontal wells and estimating resource amounts.