Abstract A good knowledge of the effective thermal conductivity (ETC) of geomaterials is required to achieve accurate study on heat transfer in a variety of earth-contact engineering applications. On account of composite-porous characteristic of geomaterials, the thermal conductivity of each component is the foundation while the porous structure is the key influence factor. Accordingly, a fractal capillary bundle model at pore scale is proposed in this paper to predict the ETC of saturated/unsaturated composite-porous geomaterials. The combination of fractal theory and capillary bundle model hypothesis is used to describe the random, disordered and extremely complicated microstructure of geomaterials. The analytical solution of the ETC is obtained by using the thermal-electrical analogy technique, which demonstrates that the proposed model is a function of variety of composition and microstructure parameters. Both the effects of these two types of parameters on the ETC of geomaterials are analyzed theoretically. Compared with other models, the present model represents specific physical meaning and shows acceptable accuracy and performance. The model predictions are further compared with the available experimental data for three types of concrete, and a good agreement between them can be found in most of saturation degrees (0.2

中文翻译：

---

复合多孔土工材料有效热导率的增强分形毛细管束模型

摘要 要准确研究各种地球接触工程应用中的传热，需要对岩土材料的有效热导率 (ETC) 有深入的了解。由于土工材料的复合多孔特性，各组分的导热系数是基础，而多孔结构是关键的影响因素。因此，本文提出了孔隙尺度的分形毛细管束模型来预测饱和/非饱和复合多孔岩土材料的ETC。分形理论与毛细管束模型假设相结合，用于描述岩土材料的随机、无序和极其复杂的微观结构。ETC的解析解是通过使用热电类比技术获得的，这表明所提出的模型是各种成分和微观结构参数的函数。从理论上分析了这两种参数对岩土材料ETC的影响。与其他模型相比，本模型代表了特定的物理意义，并显示出可接受的精度和性能。模型预测进一步与三种混凝土的可用实验数据进行了比较，在大多数饱和度（0.2