SoilVision systems Ltd. is pleased to announce the addition of full air and thermal coupling between the SVAirFlow® and SVHeat® Professional software packages. The addition extends the capabilities of the software to model flow of air in the pore space of soils. In particular, the flow of air as a result of thermal gradients can be accomplished. This addition allows benefits in modeling the following applications:
- Heap leaching: certain heap leaching operations consume oxygen during the reaction process. There is therefore the danger that the reactions will become oxygen starved and extraction will not reach capacity. Therefore air injection systems are sometimes introduced to enhance recovery. The performance of such systems can be modeled with the combined SVAirFlow® and SVHeat® software packages.
- Convective air flow in embankments: thermal convection in her embankments can now be modeled (Goering, 2003)
- Waste rock air movement: the reactions in many waste rock piles often produce heat which then affects the flow of air in such piles. The numerical modeling of such a process can aid in the long-term design of remediation scenarios for waste rock piles.
In order to examine this coupling further the following benchmark is presented. Further details of this model may be found in the SVAirFlow® Verification Manual. Goering (2003), Sun et al (2007), and Lebeau (2009) have conducted the numerical simulation of cooling an embankment using natural convection. The intent is to protect railway or roadway construction from thaw in permafrost area. This example presents the characteristics of air flow and temperature distribution as governed by natural convection within the embankment using the coupled software.
The model geometry of the right half of a symmetric roadway is shown in Figure 1, below. The embankment is constructed with ballast type of material that has a high air permeability. The pore-air in the embankment is allowed to approach equilibrium with the surrounding atmosphere. The climate temperature changes with time. The ground foundation is assumed to be uniform silt material with a high water content. On the natural ground surface, the air transfer on the boundary is assumed to impermeable, and the climate temperature varies with time. The air is assumed to be incompressible in natural convection.
The model is simulated for 21 years to eliminate the effect of the initial condition on the temperature in ground foundation. The following briefly discusses and compares the results with other software.
The patterns of convective air flow in the embankment
During winter, the cool air at the embankment surface enters into the embankment. This happens because the air density increases with the decrease in temperature. The warm air then flows out from the lower region in the embankment to the atmosphere by the process of natural convection, as illustrated in Figures 2 to 6. The onset of natural convection occurs between October and November. After that time the extensity of convective airflow increases with the decrease in temperature until January. The convective flow reduces with the increase in temperature after the end of January. During the summer, because the temperature on the embankment surface is higher than the temperature at the lower region in the embankment, a small amount of convective airflow occurs in the embankment (see Figures 7 to 9).
Compare convective flow patterns with results from others
Figure 10 is the airflow simulated from SVAirflow/SVHeat, and Figure 11 is from Goering (2003). It can be seen that the patterns of convective airflow at different times are very similar. It should be noted that the figures from Goering example are presented only as a comparison of the flow pattern. Exact comparison to the flow vectors is not possible as Goring did not published enough specific of the modeling process. As illustrated in several of the above figures, multi-airflow loops are formed in wintertime. For the two neighboring convective airflow loops, if one flows clockwise, another must move anti-clockwise. This characteristic of natural convection was also discussed in other research (Chen et al., 2008). Figure 10 further illustrates the comparison of the average value of airflow velocity at various locations (see circle points in Figure 1) with the value obtained by Goering. It should be noted that the airflow velocity for Goering in Figure 11 is the average value of airflow velocity magnitude over the whole embankment. This may be why the airflow velocity simulated in this benchmark in the SVAirFlow software has greater variation than the Goering’s value. It should be noted that the purpose of this benchmark is to replicate the reasonable flow patterns published by Goering (2003). An exact duplication of flow vectors is not needed for this benchmark to be deemed successful.