SETTLEMENT MODELLING

It is important to observe ground displacements and ground water changes as a tunnel advances. Daily analyses of these data and comparison with previous experience can lead to predictions of damage to nearby structures and modification of the construction method. The importance of ground movement observations is especially great, as they indicate changes in the quality of work and warn of damage levels in surrounding structures. This article describes the study of an approach based on neural networks (NN), used in this research for analysing and predicting behaviour during tunnelling. The neural network model was tested on a 6.5km, 9.6m-diameter length of shallow soft clayey ground tunnel for the Brasilia Metro. Three construction methods could be used for the excavation. Method A is the least expensive, and can be used if the face is stable and if there is little or no damage to nearby structures. Method B or method C may be selected if these conditions are no longer satisfied. The tests were concentrated in three locations, to investigate properties for the most typical soil layers corresponding to different geological types. The depth of ground cover and tunnel section area both affect settlement. The approach used to predict tunnel behaviour was found to be reasonably accurate and simple to use.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/0041414X
• Corporate Authors:

  Miller Freeman

  Calderwood Street
  London,   United Kingdom  SE18 6QH
• Authors:
  • Ortigao, JAR
  • Shi, Jing
• Publication Date: 1998-12

Language

• English

Media Info

• Features: References;
• Pagination: p. 30-1
• Serial:
  • Tunnels & Tunnelling International
  • Volume: 30
  • Issue Number: 12
  • Publisher: Progressive Media Markets, Ltd
  • ISSN: 0041-414X

Subject/Index Terms

• TRT Terms: Construction; Expert systems; Forecasting; Ground settlement; Groundwater; Information processing; Loss and damage; Mathematical analysis; Methodology; Quality control; Research projects; Structures; Tests; Tunnels; Warning systems
• Uncontrolled Terms: Modifications; Quality
• ITRD Terms: 6471: Analysis (math); 3352: Building (structure); 3655: Construction; 1614: Damage; 8655: Data processing; 8626: Expert system; 132: Forecast; 4383: Ground water; 9102: Method; 9048: Modification; 9063: Quality; 8557: Research project; 5792: Settlement; 6255: Test; 3374: Tunnel; 9113: Warning
• Subject Areas: Bridges and other structures; Construction; Research;

Filing Info

• Accession Number: 00761434
• Record Type: Publication
• Source Agency: Transport Research Laboratory
• Files: ITRD
• Created Date: Apr 6 1999 12:00AM