PSI - Issue 28

9

A. Prokhorov et al. / Procedia Structural Integrity 28 (2020) 1579–1589 Author name / Structural Integrity Procedia 00 (2019) 000–000

1587

10 -3

3

model BR s 1 model BR s 2 model BR s 3

2.5

2

1.5

Strain

1

0.5

0

0

0.5

1

1.5

2

2.5

Time, s

10 4

a b Figure 6. (a)Experimental results of strain measurements, (b) results of numerical simulation.

It should be noted that after the application of the thermal compensation procedure the qualitative behavior of the experimental plots did not change. The qualitative difference is difficult to explain and can have many reasons. One is that the interaction between sensors and saturated soil can change as ice content increases and residual moisture decreases, as this leads to increased adhesion between soil particles and ice. On the other hand, the kinetic equation (2) for the process of phase transition of pore moisture into ice may not be accurate enough. This assumption is supported by the fact that the segment where the experimental curves are convex downward falls on the beginning of the water freezing process. As noted above, within the segment the numerical temperature versus time curve has the greatest deviation from the readings of the thermocouple located at the center of the sample. 4.3. Ice content measurements In the one-side freezing experiment the moisture migration process can be studied. For this purpose, after full freezing the specimen was divided to eight parts. Each part of the soil was weight after cutting. Then parts was weight again after unfreezing and drying. On the basis of the difference between two measurements be it can concluded about the ice content in each part of the specimen (Fig. 7).

a b Figure 7. (a)Cut specimen, (b) Results of ice content measurements.