Effective Thermal Conductivity Estimation of Fractured Rock Masses

• Published in: Rock mechanics and rock engineering Vol. 54; no. 12; pp. 6191 - 6206
• Main Authors: Li, Zheng-Wei, Liu, Yuan, Mei, Shi-Ming, Xing, Shi-Cheng, Wang, Xiao-Kai
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.12.2021; Springer Nature B.V
• Subjects: Civil Engineering; Consistency; Contact resistance; Earth and Environmental Science; Earth Sciences; Finite element method; Fractures; Geophysics/Geodesy; Granite; Heat conductivity; Heat transfer; High level radioactive wastes; Mathematical models; Original Paper; Parameter sensitivity; Parameters; Radioactive waste disposal; Radioactive wastes; Rock masses; Rocks; Sensitivity analysis; Thermal conductivity; Thermal contact resistance; Thermal resistance; Two dimensional models; Waste disposal; Fractured rock mass; Effective thermal conductivity; Discrete fracture network; Thermal contact resistance; Estimation model
• DOI: 10.1007/s00603-021-02599-5
• ISSN: 0723-2632

In this work, effective thermal conductivity ( λ Eff ) of fractured rock masses was numerically investigated. A two-dimensional Discrete Fracture Network (DFN) model of the fractured rock masses was established based on the statistic results of natural fracture development in a potential area for high level radioactive waste disposal in China. Steady state heat transfer processes in the fractured granite rock masses were numerically simulated using finite element method (FEM). The calculated λ Eff values of the fractured granite rock masses in dry and saturated conditions are 1.99 W/(m K) and 2.31 W/(m K), respectively. Compared with the thermal conductivity of intact granite [ λ Intact , 2.5 W/(m K)], the drop rates are 20.4% and 7.6%, respectively. Sensitivity analysis was conducted on the main model parameters including fracture density ( F Density ), trace length ( F Length ), thermal contact resistance ( F TCR ), and λ Intact . The results indicate the relation between λ Eff and three fracture parameters ( F Density , F length and F TCR ) can be fitted using power law or negative exponent functions with good consistency. When fracture network parameters remain unchanged, λ Eff is in linear positive correlation to λ Intact . The slop of the fitted line is determined by the fracture network parameters. Due to the fact that distribution of generated fractures in different directions are quite uniform, λ Eff did not show significant difference in different directions. On the basis of the above-mentioned results, an estimation model was proposed for the determination of λ Eff of fractured rock masses using P 21 (total length of fracture traces per unit area), F TCR , and λ Intact . The proposed estimation model shows good consistency to the calculated results of FEM model.