Abstract Permeability prediction has been an important problem since the time of Darcy. Most approaches to solve this problem have used either idealized physical models or empirical relations. In recent years, machine learning (ML) has led to more accurate and robust, but less interpretable empirical models. Using 211 core samples collected from 12 wells in the Garn Sandstone from the North Sea, this study compared idealized physical models based on the Carman-Kozeny equation to interpretable ML models. We found that ML models trained on estimates of physical properties are more accurate than physical models. Also, the results show evidence of a threshold of about 10% volume fraction, above which pore-filling cement strongly affects permeability.

中文翻译：

---

胶结砂岩渗透率预测与基于物理和机器学习方法的比较

摘要 自达西时代以来，渗透率预测一直是一个重要的问题。大多数解决这个问题的方法要么使用理想化的物理模型，要么使用经验关系。近年来，机器学习 (ML) 带来了更准确、更稳健但可解释性较差的经验模型。本研究使用从北海 Garn 砂岩的 12 口井收集的 211 个岩心样本，将基于 Carman-Kozeny 方程的理想物理模型与可解释的 ML 模型进行了比较。我们发现在物理属性估计上训练的 ML 模型比物理模型更准确。此外，结果显示出约 10% 体积分数的阈值的证据，高于该阈值的孔隙填充水泥会强烈影响渗透率。