The physical modelling of the shear behaviour of rock joints has followed two main approaches. The first approach involves the direct shear testing of natural rock joints coupled with an empirical analysis of the shear response. The second approach has been to develop a theoretical framework for an understanding of the fundamental mechanisms of rock joint behaviour. Recent work at Monash University has attempted to combine this fundamental approach with a roughness model based on concepts of fractal geometry. This paper outlines the theory behind the fractal approach to roughness, and presents typical results of direct shear tests on rock joints prepared with so-called "fractal" surfaces.
La modelisation physique de la reaction au cisaillement de joints de roches a ete conduite de deux manières essentielles. La première a necessite un essai de cisaillement direct de la roche naturelle, allie à une analyse empirique de l'evolution de la resistance au cisaillement. La seconde a ete le developpernent d'un cadre theorique qui permette de comprendre Ie mecanisme fondamental du comportement des raccordements de roches. Des travaux recents, à l'universite de Monash, ont tente d'associer cette approche fondamentale à un modèle de rugosite etabli selon les concepts de geometrie fractale. Cet article presente la theorie de la demarche fractale à la rugosite ainsi que les resultats typiques des essais de cisaillement de joints de roches, prepares avec ce que I' on nomme des surfaces «fractales».
Die physische Representation des Scherverhaltens von Gesteinsfugen ist meistens zwei Wegen gefolgt. Einerseits verwendete man direkte Scherversuche an natuerlichen Gesteinsfugen mit einer empirischen Analyse des Scherwiderstandes: Anderseits versuchte man einen theoretischen Rahmen zu finden, der den fundamentalen Mechanismus von Gesteinsfugen erklart. Untersuchungen an der Monash Universitat haben letzlich versucht diesen zweiten Weg unter Verwendung eines Rauhigkeitsmodells abgeleitet von Begriffen der Fraktalen Geometrie zu verfolgen. Diese Abhandlung gibt die Umrisse der Theorie fuer die fraktale Behandlung von Rauhigkeit und vergleicht typische Resultate von direkten Scherversuchen, die an Gesteinsfugen mit sogenannten "fraktalen" Oberflachen durchgefuehrt wurden.
The physical modelling of the shear behaviour of rock joints has followed two main approaches. The first approach involves the direct shear testing of natural rock joints coupled with an empirical analysis of the shear response (Barton and Choubey, 1977; Bandis, 1990). In such approaches, the response of the joint is understood in a global sense, without necessarily understanding the mechanics or principles of the shear strength development. The joint is largely a "black box" and empirical relationships are determined on the basis of the outputs from a large number of tests where a range of inputs, such as joint roughness, length, intact strength and normal stress are selectively varied. Although the shear displacement response is not directly predicted by such approaches, the shear stiffness may be determined from further empirical relationships. These approaches have been very successful and are widely adopted in rock mechanics practice, however, their use is compromised by the possibility of in-situ conditions being outside the range of tested input parameters, or the interaction of input parameters not being appreciated in the development of the models. The second approach has been to develop a theoretical framework for an understanding of the fundamental mechanisms of rock joint behaviour (patton, 1966;'Ladanyi and Archambault, 1970; and Johnston and Lam, 1989). The development of such fundamental approaches is vital to the advancement of knowledge of rock joint behaviour, however, the models developed have been based primarily on the analysis of joints represented by simple triangular asperities.
