Numerical simulation of compressed air tunnel advance

Compressed air in shallow tunnelling is used to counteract the tendency of water to flow into the opened space. The method of compressed air has gained new importance recently, because in urban areas it does not influence the surrounding groundwater. Another advantage is that the transportation of existing contaminants into the direction of the tunnel can be prevented. One major disadvantage is that the excavation work takes place under severe conditions with respect to the health of the tunnellers. The amount of air flow varies depending on the soil conditions, the geometry of the tunnel and the applied air pressure. Air flow takes place both through the unsupported tunnel face and shrinkage cracks of the shotcrete lining. The required air pressure must be controlled and adjusted if necessary continuously. During tunnelling, advance air flows into the surrounding soil. The primary fully saturated soil becomes partially saturated. This has large influence on the amount of air flowing not only through the tunnel face but also through cracks in the shotcrete lining. While the amount of air flow through the tunnel face is relatively constant, the amount of air flow through the shrinkage cracks increases with the length of the excavated tunnel section and becomes the governing factor. A large-scale laboratory test to simulate the air flow through the shotcrete lining and the soil was conducted. Different parameters have been varied. These parameters were: (1) the amount of air pressure; (2) the width of the cracks in the lining; (3) the thickness of the lining and (4) the soil type. Subsequently a numerical model was developed that allows the simulation of the flow behaviour in the laboratory test and further extended to simulations of compressed air tunnel advance. TOUGH2 was chosen to perform the numerical analyses. To compute the air flow through the tunnel face and the shotcrete lining a three-dimensional model was developed including the progress in tunnel advance. To model the tunnel advance elements were sequentially removed in the tunnel face area. The model considered the relative permeability saturation dependency of a soil. In the present contribution two models with varying tunnel advance rates were compared. The reduction of the total amount of air flow due to an increased tunnel advance rate was demonstrated. The model with a tunnel advance rate of v = 1 m/d demands about 80% more air per hour than a model with an advance rate of v = 3 m/d. (A). "Reprinted with permission from Elsevier". For the covering abstract see ITRD E124500.

  • Availability:
  • Authors:
    • SCHEID, Y
  • Publication Date: 2004-7


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01011625
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • Files: ITRD
  • Created Date: Dec 19 2005 3:21PM