GEOMECHANICAL EVALUATION OF TUNNEL STABILITY UNDER FAILING ROCK CONDITIONS IN A HIMALAYAN TUNNEL

Abstract

Rock pressures and displacements play a dominant role in determining the stability of a tunnel opening. The search for a reasonably accurate method for predicting the rock pressures and displacements had, therefore, been an important issue since decades. Rock pressures and displacements depend upon the behaviour of tunnelling media which is difficult to predict. Hence, the pre-assessment of rock pressures and displacements is a complicated task. The current theories in vogue for assessment of the rock pressures are either based upon empirical approaches or on analytical methods. Their efficacy has not been adequately proved by field studies. In the present study, an attempt has been made to estimate rock pressures and displacements with the help of current theories and the results have been compared with the observed data. The basic geomechanical data was collected at the project sites through investigations and instrumentation. The predicted lithology along the tunnel alignment proved inaccurate and consequently the pre-assessment of rock pressures was also faulty leading to failure of tunnel supports, The behaviour of tunnel supports was affected adversely by the presence of major and minor geological defects. The rock pressures estimated with the help of empirical approaches of Terzaghi(l%6), Deere et al (I969) and Barton et al (1974) were nearer to the observed pressures in the case -IVof non-squeezing rocks only. The analytical approach, based upon the elastoplastic behaviour of failed rock mass within broken zone has, there fore, been used to assess the rock pressures in squeezing rock conditions which were prevalent in the tunnels under reference. In the published literature, no method existed for the analysis of tunnel instrumentation data. Hence a graphical method has been evolved for estimation of the radius of the broken zone. Labasse's (1949) equation, modified suitably, had been used to study the behaviour of the broken zone and it was found out that the coefficient of volumetric expansion (k) was not constant throughout the broken zone as proposed by him. Further, it was found to vary with time and a trend of gradual decrease in its value with radial distance was perceptible. The peak values of overall k for different rocks were between 0.002 to 0.009. The elastoplastic theory was modified for non-hydrostatic stress field and approximate closed-form solutions were obtained to derive the ground reaction curve. Methods given in the published literature were used to assess the strength para meters of rock mass and the primitive stresses were assumed judiciously. The actual support reaction curves obtained from the field instrumentation data were superposed on the ground reaction curves for the respective cases and a criterion has been proposed for the estimation of equilibrium pressures and displacements. The estimated rock pressures were no better -vthan those obtained through Barton's approach. However, the elastoplastic theory has been found to be an important tool for the study of the mechanism of rock-support interaction in squeezing rock conditions. The results tend to support Heim's hypothesis of hydrostatic primitive stress field. The steel rib supports in the Giri tunnel failed to withstand ultimate rock pressures and consequently, suffered severe twisting and buckling. However, there was no indication of an impending collapse. This situation was possibly due to the reduction of rock pressures with increasing tunnel-wall displacements, a logical conclusion derived from the elasto plastic theory.