This paper presents characteristic features of rock behaviour during excavation obtained through field measurements at many sites for underground power station caverns. Some findings are as follows:

  1. Joint openings playa great role in rock displacement.

  2. Stresses in arched concrete linings do not depend on horizontal stress of ground pressure, but on the horizontal displacements of cavern walls.

  3. The degraded states of relaxed zones are observed by means of the elastic wave velocity test, the Lugeon test, and the bore-hole television method.


Cette communication presente les caractèristiques de comportement des roches lors d'excavations et obtenues par des mesures in situ relevees dans de nombreuses galeries de centrales souterraines. En voici quelques resultats: 1) L'ouverture des joints joue un rôle important dans le deplacement des roches. 2) L'effort dans le revêtement de beton de l'arc ne depend pas de l'effort horizontal de la pression du sol mais du deplacement horizontal de la paroi de la galerie. 3) L'etat d'alteration des zones de decompression est observe par essais de velocite des ondes elastiques de Lugeon et par introduction d'une camera de television dans les sondages.


Diese Abhandlung prasentiert die charakteristischen Eigenschaften des Felsverhaltens wahrend des Ausbruchs, die durch in-situ Messungen an vielen Baustellen fuer Kavernen fuer unterirdische Kraftwerke erhalten wurden. Nachfolgend werden einige dieser Charakteristiken angefuehrt:

  1. Kluftöffnungen spielen eine große Rolle bei der Gebirgsverformung.

  2. Die Spannungen in den Betonauskleidungen hangen nicht von der horizontalen Spannung des primaren Spannungsfeldes, sondern Von der horizontalen Verdrangung der Kavernenwande ab.

  3. Der degradierte Zustand der entspannten Zonen wurde durch elastischeWellengeschwindigkeitspruefung, Lugeon-Pruefung und Bohrlochfernsehmethode nachgewiesen.


A lot of large caverns for underground power stations have been excavated in Japan. The authors have been developing one method of numerical analysis to render these excavation sites as safe as possible (Hayashi et al., 1970). Comparison between forecasting by the method and measured results has proved the method to be fairly reliable (Hibino et al., 1977). Presented in this paper are the characteristics of rock behaviour obtained through the measurements:

  1. the different characteristics of displacements in ceiling rocks and cavern walls,

  2. the typical behaviour of joint openings.

  3. the dependency of stresses in an arched concrete lining on convergences of cavern walls

  4. the degradation of relaxed zones.


Geological conditions, magnitudes of ground pressure, sizes of the caverns and the other conditions relating to the stability of caverns are listed in Table I. The vertical stresses of ground pressure range from 1.9 to 10.8 MPa, which means that the overlying rock depths are 80~480 m. At each site the following in situ tests have been carried out to obtain mechanical properties of rock foundation. These include jack tests for deformability Eo, rock shearing tests for shearing strength τo, creep tests for creep factor α, and stress relief method (overcoring method) for ground pressure. In advance of excavation, forecasting of rock behaviour around cavities had been performed with the numerical analysis method developed by the authors. During the excavation works, measurements of rock behaviour around the caverns were performed for the purpose of safety during construction and to ensure the validity of the numerical analysis. The main kinds of measurements were as follows: deformations of rock foundation (subsidences of ceiling rocks and horizontal displacements of walls of the caverns), stresses in arched concrete linings, and variation of stresses in the strands of reinforcement.


Displacements are usually measured with rock displacement meters. In order to detect the sizes of relaxed zones, the distribution of displacements should be cleared so that the measuring lengths of the displacement meters, which are fixed at the excavation surface, will often be 1, 3,5,10,15 m, etc. 3–1 Subsidences of ceiling rock masses A typical example of time history for subsidence in ceiling rocks is shown in Fig. 1 (Hibino et al., 1979). Observed results at the other sites have the same tendencies as those of Fig. 1. Fig. 2 gives the distributions of subsidences at the completion of excavation at several sites. From these Figs., the following findings are clear:

  1. Almost all of the total subsidence was generated at the stage of the arch part excavation, and the following excavation of the main part did not increase any subsidence; on the contrary, there was a tendency toward decreasing of the settlement. In Fig. 1 the results of the forecast by the numerical analysis are also shown, and rather good agreement between the measured and calculated settlement was found in both total values and time history.

  2. The largest values of the total measured subsidences occurred in a limited part of the rock surface in a region of only 0 ~ 5 m (Fig. 2).

This content is only available via PDF.
You can access this article if you purchase or spend a download.