Numerical Simulation Of Artificial Ground Freezing In Tunneling Subjected To Seepage Flow

The future development of society and the continuous growth of population are creating the demand for greater utilization of underground space in urban areas.Excavation support is required to build underground structures.For the construction of underground structures,the soil should be strong enough so that it can bear the overburden loads easily during excavation.In the water-rich soft ground,it is extremely difficult to excavate the soil for underground structures without strengthening the ground.Artificial ground freezing(AGF)is a commonly used technique in geotechnical engineering for ground improvement such as slope stabilization,groundwater control and temporary excavation support during tunnel construction in soft ground with high groundwater levels.It is a versatile technique that increases the strength of the loose ground and makes it impervious to water seepage.Excavation can proceed safely inside the frozen ground structure until construction of the final lining provides permanent support.In contrast to grouting works the freezing method is completely reversible and has no environmental impact.Sometimes for excavation to build underground structures,the soil is found to have water seepage.The strength and stiffness properties of frozen soil and the time of establishment of a complete frozen soil body with full temporary load carrying capacity are considerably influenced by groundwater or seepage flow since it provides a continuous source of heat.In case of large seepage flows,a state of thermal equilibrium can be reached,in which freezing stops and the closure of desired frost wall cannot be developed.Based on the above issues,this thesis “Numerical simulation of artificial ground freezing in tunneling subjected to seepage flow” has been carried out by using COMSOL Multiphysics Software.Based on the heat transfer and seepage theory in porous media a 2D Finite element tunnel model has been established to simulate the variation of temperature field with the passage of time.The tunnel model around which freezing arch is required is subjected to horizontal groundwater flow normal to the axis of the tunnel.The effect of groundwater flow on the freezing arch has been carried out at different water seepage velocities at constantly increasing time.Finally,the simulation results of temperature distribution have been compared at different water seepage velocities at constantly increasing time.It is found that in the absence of seepage flow,there is a symmetrical distribution of temperature around the freezing pipes.But in the presence of seepage flow there is an unsymmetrical distribution of temperature around the freezing pipe,the upstream side of the freezing pipe is more affected by seepage flow as compared to the downstream side.It has been found that at a specific seepage velocity the freezing is not possible.The simulation results have been validated by comparing it with past experimental data.It has been found that there is a good agreement between numerical results and the past experimental results.In addition,the behavior of different soil types against artificial ground freezing has been studied.The different soil type has been selected and simulated.The behavior of temperature distribution with the time of different soil has been carried out.In the end,the simulation results of the temperature distribution of different soil types have been compared.It is found that the sandy soil is more favorable to artificial ground freezing as compared to clay soil.The numerical simulation of artificial ground freezing for Harbin metro station has been carried out by using COMSOL Multiphysics software.A 2D Finite element model has been established.The temperature distribution at required monitoring points has been collected by numerical simulation.Finally,the in-situ results of the development of temperature with time have been compared with simulated results.It has been found that there is a good agreement between in-situ and numerical simulated results.