1 INTRODUCTION
Rock bolts have been recognized as one of the most effective support measures for stabilizing jointed rock masses around tunnels. As far as the author is concerned, however, there seems to be no reliable design method presently available for dimensioning the rock bolt support system. This may be mainly due to the fact that the function mechanism of rock bolts installed in jointed rock masses has not yet been fully investigated. Therefore, in practice, the design of rock bolts is based on the experience of and judgments by well-trained engineers. 1) -4) Engineers always run into difficulties, however, and are faced with the question of how to determine the number and length of the rock bolts necessary to stabilize the tunnels.
The purpose of this study is to give an answer to the above question by proposing a method which is capable of both evaluating the effectiveness of rock bolts quantitatively, and determining the optimal number and length of them. The method is based on displacement measurements carried out during the course of tunnel excavations. The effectiveness of rock bolts is assessed by the strain occurring in the surrounding rock masses, and the optimal number and length of the rock bolts are determined by controlling the strain so as to maintain it within the level of critical strain.
2 MONITORING THE STABILITY OF TUNNELS
The stability of tunnels during and after construction can be monitored in terms of either the stress or strain occurring in the rock masses around the tunnels. It is extremely difficult in practice, however, to measure stress, and the results are not reliable enough. The measurement of strain, on the other hand, can be done without difficulty, because strain is kinematically related to displacements which are easily measured by various types of extensometers and convergence meters, which are commercially available. The results of displacement measurements are generally reliable. Sakurai (1981) proposed a method for determining strain from measured displacements using the kinematic relationship alone. Sakurai and Takeuchi (1983) also proposed a more sophisticated method based on a back analysis.
When strain is to be determined, it is recommended to monitor the stability of tunnels by evaluating the strain directly, instead of transforming it into stress. This is not only because Young's modulus is unnecessary, but also because the determination of the allowable strain of rock masses is much easier than that of the allowable stress (see Fig. 1).
Fig. 1 Monitoring of underground openings based on displacement Measurements (available in full paper)
Sakurai (1981) proposed a strain-based monitoring method which has been named the "Direct Strain Evaluation Technique (DSET)". According to this method, the stability of tunnels can be assessed by comparing the strain occurring in the rock masses with the allowable strain of the materials. If the occurring strain is small enough to be within the value of the allowable strain, tunnel stability may be maintained. If the occurring strain tends to exceed the allowable strain, the stability is questionable, and the installation of rock bolts is recommended.
