Rock joint coefficients and their computerized classification

Abstract

Abstract A computerized method for determining rock joint coefficients is presented. Two relative similarity indicators are introduced to classify surface morphology of rock joints. The classification enables to compare investigated and database rock joints. Such a comparison aims at finding the couple of surfaces that are distinguished by the highest dynamical conformity. The first absolute indicator results from the Fourier matrix and evaluates wavy shapes of surfaces. The second absolute indicator quantifies the heights of surface reliefs and is defined as the root mean square height of the surface outline. Numerical reliability of these indicators is tested within the surface analysis of a series of limestone specimens. Besides the computerized assessment, 25 people have performed visual assessment of these limestone specimens. The results of visual assessments have been statistically processed and compared to the results received from the computerized procedure. The newly introduced absolute indicators have proved to be prospective numerical tools for evaluating joint rock coefficients.