Study on the evaluation method of heat transfer characteristics of rocks with cracks for the development of the deep ground.

开发了一套个人计算机系统和一套热像图采集程序，用于分析地下热流的室内实验，并在此基础上，利用小型模型地面，研究了孔隙水饱和多孔介质中的传热问题。该试验装置可构造不同渗透率的地层，通过改变水头可获得任意量的地下水流量。并编制了有限元程序，对实验结果进行了分析，计算了各组分的热常数。对比理论结果和实验结果，研究了水流对土壤表观导热系数的影响，发现孔隙水对流对土壤整体导热系数有较强的影响，由De Vries提出的经验方程给出了土壤导热系数的准确值。去年对渗流理论和等效包涵法在土壤热常数估算中的应用进行了研究，并得出结论：由于理论估算中存在结构假设和平均过程导致的误差，这些理论并不总是便于计算土壤热常数。此外，与这两种理论假设的估计值相比，所提出的有限元计算方法给出了足够准确的结果。因此，利用渗流理论和等效包含方法进行的分析性研究被排除在今年的研究之外。

A personal computer system and a program to capture thermographs was developed for the analyses of laboratory experiment of the heat flow in the ground and then the heat transfer in the porous media saturated by pore water was examined using a small model ground. In this experimental equipment the ground with different permeability can be constructed and any amount of ground water flow can be given by changing the water head. A FEM program was also developed to analyze the experimental results and to evaluate the thermal constants of the constituents. Comparing the theoretical results and experimental ones, effect of water flow to the apparent heat conductivity of soil was studied, and it was found that there are strong effect of convection of heat by pore water on the heat conductivity of soil as a whole and an empirical equation proposed by De Vries gives an accurate values of heat conductivity of the soils.Application of percolation theory and equivalent inclusion method to the estimation of thermal constants of soils were examined last year and it was already concluded that these theories are not always convenient to evaluate the thermal constants of soils because the theoretical estimation always contains errors caused by assumption of structures and averaging procedures. In addition to that it was found that the proposed method of FEM calculation gives enough accurate results compared to the values estimated from these two theoretical assumptions. Therefore analytical studies by use of percolation theory and equivalent inclusion method was excluded from the studies of this year.