Abstract: Accurate prediction of pore pressure in carbonate formations is a critical prerequisite for maintaining wellbore stability and ensuring the efficient exploration and development of carbonate oil and gas reservoirs. However, carbonate formations are characterized with complex geological conditions, strong heterogeneity, and significant regional variation, which leads to great difficulty in achieving accurate pore pressure prediction, as well as the low applicability of prediction models. There are still significant challenges in enhancing prediction accuracy and expanding model applicability. To deepen the scientific understanding of pore pressure prediction methods for carbonate formations, this paper explores the formation mechanisms of abnormal pressure, classifies and compares the main prediction methods for the current pressure in carbonate formations, and systematically summarizes the principles, advantages, and disadvantages of each method. Using the marine carbonate formations in the southern Sichuan Basin as a case study, the adaptability of different prediction methods were comparatively evaluated. Key findings include:(1)The prediction methods such as the equivalent depth method and Eaton method, which are built based on undercompaction theory, rely on establishing reasonable compaction trend lines. These methods exhibit low accuracy and poor applicability for carbonate formations. (2)Numerical simulation-based methods can accurately describe formation geometry and obtain high-precision numerical solutions through fine grid division and high-order finite element parameters. However, their stringent parameter requirements and high computational complexity limit practical implementation. (3)As for the methods established based on statistical analysis and artificial intelligence algorithms, prediction models are obtained by analyzing large volumes of regional data. However, model variability across different blocks limits their applicability. (4)The rock physics-based prediction methods for pore pressure include flow-solid coupling models and wavelet transform-based models, in which rock physics models are constructed to determine the response relationship between pore pressure and elastic wave parameters of carbonate formations. However, their high complexity and parameter-processing difficulties hinder widespread application. Finally, based on the comprehensive understanding, the paper outlines future research direction for pore pressure prediction in carbonate formations, providing theoretical guidance and technical support for the safe and efficient exploration and development of oil and gas resources in carbonate reservoirs.

Key words: carbonate rock, abnormal formation pore pressure, prediction methods, formation mechanism, rock physics model

摘要： 碳酸盐岩地层孔隙压力准确预测是维持碳酸盐岩地层井壁稳定,保障碳酸盐岩油气藏高效勘探开发的重要前提。碳酸盐岩地层地质条件复杂、非均质性强且区域差异特征明显,导致其地层孔隙压力准确预测难度高、预测模型普适性差。提高碳酸盐岩地层孔隙压力预测精度,拓展预测模型适用性仍面临极大挑战。为深化碳酸盐岩地层孔隙压力预测方法的科学认识,对碳酸盐岩异常压力形成机制进行讨论,针对当前碳酸盐岩地层孔隙压力预测主要方法进行分类对比,系统总结了各方法的原理与优缺点,并以四川盆地南部地区海相碳酸盐岩地层为例,开展不同预测方法的适应性对比评价。研究结果表明:①基于欠压实理论的预测方法,如等效深度法、Eaton法,依赖于建立合理压实趋势线,对碳酸盐岩地层孔隙压力预测准确性低,适用性较差;②基于数值模拟预测方法能够精细地描述地层的几何特征,通过精细网格划分和高阶有限元参数可获取高精度数值解,但参数获取需求高,计算复杂度高;③基于数理统计与人工智能算法的预测方法通过对研究区块大量相关数据处理分析获取预测模型,模型区块差异性明显,应用场景局限性较大;④基于岩石物理学的地层孔隙压力预测方法,包括流固耦合预测模型、基于小波变换的预测模型等,通过构建岩石物理模型,获取碳酸盐岩地层孔隙压力对弹性波参数的响应关系,对地层孔隙压力进行高精度预测,但模型复杂、参数处理难度高。最后,对碳酸盐岩地层孔隙压力预测的发展趋势进行了展望。

关键词: 碳酸盐岩, 异常地层孔隙压力, 预测方法, 成因机制, 岩石物理模型