Numerical Simulation of Tunnel Advance using the BEM

The aim of the project is to develop a novel software for the analysis of the displacements and stresses that occur due tunnel construction. Particular emphasis is placed on the new Austrian tunneling method (NATM), where tunnel advance is by blasting and the primary means of support is shotcrete and rock bolts. The software is based on the Boundary Element method (BEM), whose capabilities have been extended in order to allow the consideration of sequential excavation, the installation of ground support and visco-plastic material behaviour. The boundary element method has the advantage over the more commonly used finite element method (FEM) in that no mesh truncation is required and that the number of elements which needs to be used is reduced by an order of magnitude. Boundary element meshes are also easier to generate and the solution is faster. The novel approch in this project, which makes the software extremely user friendly, is the adaptive generation of internal cells for the integration of the volume integrals that occur when dealing with plasticity in the BEM. In addition a facilty is implemented that allows to simulate the sequential excavation/construction processes that occur in NATM tunnelling. The resulting analysis software is intended to significantly reduce the effort for the three-dimensional numerical simulation of tunnel advance. As a consequence consulting bureaus will be able to perform these analysis on a routine basis. This is expected to contribute to the increase the safety of tunnels.

The results of the project will make a significant contribution to the development of innovative software that will enable tunnel engineers to perform numerical simulations at the tunnel site in almost real time with a minimum of effort. Such simulations allow an engineer to test different designs by "constructing a tunnel in the computer" in a similar way as for example the crashworthiness of a car is tested today. This software tool would assist in finding the most optimal way of constructing a tunnel. The main aim is to make such simulations as easy to use and as realistic as possible. Currently used simulation programs are too complicated to use and sometimes do not give very realistic results. The scientific advances made in this project have a direct impact on the work performed by the institute for structural analysis in the European integrated project "Technology Innovation in Underground Construction" (6 th framework programme) whose aim it is to develop a fast, accurate and efficient simulation tool for the design of tunnels. It is expected that this development will have significant impact on the cost and speed of the construction of tunnels. It also may contribute to solving many traffic and pollution problems that affect society, since most of the traffic can be moved underground. Numerical methods, which are currently used in tunnelling such as the Finite Element (FEM) and Finite Difference Methods, have major drawbacks with respect to user friendliness and efficiency. The current state of the art is that a 3-D simulation of tunnel construction is performed by specialists and takes considerable effort. A powerful alternative to these methods is the Boundary Element Method (BEM). However, this method is not very well known and the scientific community working in this field is small. Therefore the development of the method is far behind the FEM. The main advances in the project therefore relate to the further development to the BEM to extend its range of application to tunnelling. The scientific outcome of the project is also applicable to other applications in engineering, which involve non- linear, time-dependent material behaviour and continuous small changes in geometry.