Abstract

As an important parameter to consider the thermal stability of road engineering in cold regions, thermal conductivity is largely affected by temperature, water content, dry density, and etc.. When investigating thermal conductivity, setting the dry densities and water contents of soil samples according to compaction curve agrees better with actual engineering conditions because of its compaction characteristic for subgrade soil. Therefore, experiments on thermal conductivity of 9 groups of soil samples corresponding to different water contents at the compaction curve were carried out under −40 ℃~20 ℃. It was found that thermal conductivity decreases linearly with the decrease of temperature in positive temperature range and increases exponentially in negative temperature range, which is closely connected to the variations of thermal conductivity, contents and interactions of every phase with temperature. Multiple linear regression analysis showed that water content has a slightly higher effect on thermal conductivity at positive temperature than that of dry density; however, its effect is about twice of that of dry density at negative temperature. Then, an assessment method combining Root Mean Square Error (RMSE) and Root Mean Square Value (Xrms) was proposed to evaluate the predictive capabilities of two classical models (Johansen’s model and the CK model) effectively.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call