Temperature effects and prediction model of thermal conductivity of soil

Considering the influences of environmental temperature on thermal conductivity of soil is necessary for the optimization design and safety assessment of underground thermal engineering projects. The thermal conductivities of lateritic clay, silt clay, soft clay and bentonite at different temperatures are measured by using the thermal probe method, and the temperature effects of thermal conductivity of soil and its influencing factors are analyzed. A weighted geometric average model considering the temperature effects of thermal conductivity of soil is then established, and is compared with the traditional predictive models. The test results show that the thermal conductivity of soil increases with the increase of temperature, and its temperature effects decrease with the increase of dry density. The temperature has a great influence on the thermal conductivity of unsaturated soil, but a weak influence on the thermal conductivity of dry and saturated soil. The temperature effects of thermal conductivity of soil may depend on the change of the latent heat transfer of vapor. The more the moisture and vapor migration channels that can provide for the latent heat transfer of vapor in soil, the more significant the temperature effects of thermal conductivity of soil. The calculated results show that the proposed weighted geometric average model provides the best fitting to the measured data against the three other traditional models, and can predict well the influences of water content and dry density on the temperature effects of thermal conductivity of soil, while the prediction accuracy of the Tarnawski model, Gori model and Leong model is lower than that of the weighted geometric average model.