Ju-Hua Li, Xiao-Qin Zeng, Cui-Hao Lian, Hai Lin, Shiduo Liu, Fengyu Lei, Youyu Wan
Hubei Key Laboratory of Oil and Gas Drilling and Production Engineering (Yangtze University), Wuhan, 430100, Hubei Province, China.
School of Petroleum Engineering, Yangtze University: National Engineering Research Center for Oil & Gas Drilling and Completion Technology, Wuhan City, 430100, Hubei, China.
Sci Rep. 2025 Mar 18;15(1):9318. doi: 10.1038/s41598-025-93224-w.
The traditional logging evaluation of comprehensive sweet spots in shale oil reservoirs has problems such as complex explanatory parameters, incompatible quantitative characterization scales, and low-cost efficiency. A method based on the fractal characteristics of conventional logging curves is proposed to evaluate the comprehensive sweet spots of fractured horizontal wells in shale oil reservoirs. Firstly, the existing evaluation parameters and methods were reviewed, pointing out the limitations of traditional logging evaluation methods. Furthermore, we analyzed 63 fractured sections from three horizontal fractured wells in the Yingxiongling shale oil reservoir of the Qinghai Oilfield, using tracer monitoring data. By applying wavelet transform to reduce noise in high-frequency signals from conventional logging curves, we then used multifractal spectrum analysis and R/S analysis to extract the multifractal spectrum width (∆α) and fractal dimension (D) from four conventional logging attributes: natural gamma logging (GR), acoustic time difference logging (AC), density logging (DEN), and neutron logging (CNL). A multi-attribute comprehensive fractal evaluation index was developed by using the post-fracturing tracer monitoring profile as a constraint and applying the grey relational analysis method. This approach enabled a quantitative classification and evaluation of the key sweet spots in shale oil reservoirs after fracturing. The results show that the comprehensive fractal evaluation index of the high-yield well section after Class I layering is 0.75<∆ α'<1, 0 < D'<0.25; 0.35<∆ α'<0.75, 0.25 < D'<0.8 in the middle well section of Class II layer; Class III low production well Sect. 0<∆ α'<0.35, 0.8<∆ α'<1. Finally, a prediction model for physical property parameters characterized by fractals was introduced using machine learning algorithms, which is 31.9% more accurate than the conventional interpretation physical property parameter prediction model for the comprehensive sweet spot of fracturing. This evaluation method is a concise approach to comprehensively evaluate the sweet spot area based on the extraction of multifractal spectral characteristic parameters from conventional logging data. It is of great significance for characterizing the volume fracturing effect of shale oil and providing technical support for the effective development of shale on a large scale.
页岩油藏综合甜点区传统测井评价存在解释参数复杂、定量表征尺度不兼容、成本效率低等问题。提出一种基于常规测井曲线分形特征的方法来评价页岩油藏压裂水平井的综合甜点区。首先，回顾了现有的评价参数和方法，指出了传统测井评价方法的局限性。此外，利用示踪剂监测数据，对青海油田英雄岭页岩油藏3口水平压裂井的63个压裂段进行了分析。通过应用小波变换降低常规测井曲线高频信号中的噪声，然后利用多重分形谱分析和R/S分析从自然伽马测井（GR）、声波时差测井（AC）、密度测井（DEN）和中子测井（CNL）这4种常规测井属性中提取多重分形谱宽度（∆α）和分形维数（D）。以压裂后示踪剂监测剖面为约束，应用灰色关联分析方法，建立了多属性综合分形评价指标。该方法实现了对页岩油藏压裂后关键甜点区的定量分类与评价。结果表明，Ⅰ类分层高产井段综合分形评价指标为0.75<∆α'<1，0<D'<0.25；Ⅱ类层中段井段为0.35<∆α'<0.75，0.25<D'<0.8；Ⅲ类低产井段为0<∆α'<0.35，0.8<∆α'<1。最后，引入了基于机器学习算法的分形表征物性参数预测模型，该模型对压裂综合甜点区的解释比传统解释物性参数预测模型的准确率提高了31.9%。这种评价方法是一种基于从常规测井数据中提取多重分形谱特征参数来综合评价甜点区的简洁方法。它对于表征页岩油体积压裂效果、为页岩大规模有效开发提供技术支持具有重要意义。
