The time-dependent behaviour of small tunnels in highly stressed, jointed rock tested during laboratory tests is described after a brief description of the test apparatus and some typical test results. For the tunnels tested it was possible to determine unique tunnel convergence rate functions that separate three modes of tunnel behavior:
prefailure mode,
stable yield zone propagation mode, and
unstable or rupture mode.
On the bases of these convergence rate functions the performance of tunnels excavated in the same rock material was predicted based on convergence measurements. The practical implications of these convergence rate functions are discussed.
Après avoir donne une brève description des appareils d'experimentation utilises et presente quelques resultats de tests typiques, on decrit le comportement, en fonction du temps, de petits tunnels excaves dans une roche fissuree soumise à de fortes contraintes et etudies au cours d'experiences en laboratoire. On a pu determiner des fonctions de convergence propres à trois modes de comportement des tunnels, à savoir: a) mode d'avant rupture; b) mode stable de propagation de la zone de rendement et c) mode instable ou de rupture. En se basant sur ces fonctions de taux de convergence on a pu predire la performance de tunnels dans un même type de roche. Enfin, on examine les implications pratiques de ces fonctions de taux de convergence.
Das zeitabhangige Verhalten von Tunneln in ueberbeanspruchtem, geklueftetem Gebirge wird beschrieben zusammen mit einer kurzen Erlauterung der Versuchsanlage und typischen Versuchsresultaten. Fuer die Tunnel, die im labor geprueft wurden, konnten Konvergenz-Geschwindigkeits-Funktionenbestimmt werden, die es erlauben, drei Tunnelverhaltensgruppen zu differenzieren:
Verhalten vor dem Bruch,
Verhalten wahrend beschrankter, und
wahrend unbegrenzter Ausbreitung der Bruchzone.
Mit Hilfe dieser Konvergenz-Geschwindigkeits-Funktionenwar es möglich, das Verhalten von Tunneln in ahnlichen Gesteinsmaterialien vorherzusagen. Die praktische Bedeutung dieser Konvergenz-Geschwindigkeits-Funktionen wird kurz erlautert.
The success of modern tunnelling technologies such as mechanized excavation by tunnel boring machines or the New Austrian Tunnelling Method (NATM) depends largely on the interpretation of the time-convergence relationship that is controlled by the time dependent rock mass behaviour and the excavation - support history, particularly when excavating in incompetent rock. The stand-up-time and the rate of tunnel closure are dominated by the excavation method and sequence, the support system and the installation procedure. The time-dependent tunnel wall convergence is initially controlled by the rate of face advance and then by the time-dependent stress redistribution processes resulting from softening or yielding ahead of the tunnel face and near the tunnel walls. Far from the tunnel face the time-dependent behaviour is controlled largely by the creep properties of the rock mass. For the purpose of studying these time dependent processes controlling the performance of tunnels in highly stressed or overstressed rock masses, the tunnel convergence and the internal rock mass strains near the tunnel wall. were monitored during process simulation tests (PST) in the laboratory. The instantaneous and time-dependent rock mass response was observed while a circular tunnel was excavated, as well as during creep stages following tunnel excavation, in a jointed rock mass with time dependent strength and deformation properties. The total accumulated tunnel closure or rock mass strain can seldom be determined accurately in the field and the total time dependent displacements are difficult to determine because the initial strain rates are high when compared with the loading excavation rate. Moreover, access for strain measurements ahead of the tunnel face is often limited or impossible. In practice, it has been found convenient to use the deformation rate rather than the displacements as an indicator to evaluate tunnel performance. For example, at the Arlberg Tunnel (John (1981)) the tunnel wall displacement rate was used to determine when additional rock bolting was required and what the optimum bolt length should be. Additional or longer bolts were installed when the deformation rate exceeded 50 mm/day, and reasonable development of loads on the final lining was observed if the residual tunnel wall deformation rate in a tunnel with a diameter of 11 m was equal to or less than 10 mm/month, [John (1977)). These values correspond approximately to tunnel closure rates of 37 × 10-3%/hour and 0.25 × 10-3%/hour respectively. Because of the practical significance of these closure or strain rates most of the following data interpretation will be based on the time-dependent observations presented in double logarithmic deformation rate vs time plots. This relationship was found to be nearly linear for the test material (coal) [Kaiser et al. (1981)] and a similar linear relationship was observed from the tunnel closure and the radial rock mass strain observations [Kaiser et al. (1981). As discussed by Kaiser (1981) and Kaiser and Morgenstern (1982) tunnel closure measurements reflect the overall behaviour of the underground opening but must be combined, with extensometer measurements to delineate the yield or failure mechanisms.
