| Accurate quantification of roughness is important in modeling strength, deformability and fluid flow behaviors of joints. Both stationary and non-stationary fractional Brownian profiles with known values of the fractal dimension and other profile properties were generated. The generated profiles were in general quite similar to the roughness profiles of natural rock joints. For different values of the input parameter of the line scaling, variogram and roughness-length methods, D was calculated for the generated self-affine profiles to investigate the accuracy of each method. The results obtained are used to suggest guidelines to calculate fractal-based roughness parameters accurately for natural rock joint profiles.; It is important to know the strength of rock discontinuities in different directions in dealing with rock engineering systems which are subjected to various external forces. The roughness of a natural rock joint was measured in different directions using a laser profilometer. A model material which is a mixture of plaster of Paris, sand and water was used to make model material replicas of the natural rock joint. Direct shear tests were performed on rough model material joints at five different normal stresses, in the same directions which were used for the roughness measurements. The scratched areas on rough joints resulting from direct shear tests were used along with the joint surface topography to estimate the asperity shear area of the joints. Direct shear tests were conducted on smooth model material joints to estimate the basic friction angle. Required tests were conducted on intact model material to develop a peak shear strength criterion for the intact model material. The direct shear data obtained on the rough joint for 3 of the 5 normal stresses were used along with other laboratory test results to develop a new peak shear strength criterion which includes two fractal based stationary roughness parameters, a non-stationary roughness parameter, basic friction angle, normal stress, joint compressive strength, intact shear strength and area of asperity shear as a proportion. The developed criterion was used to predict the peak shear strength in different directions at the other two normal stresses which were not used in developing the peak shear strength model. The comparison between the predictions and the measured values showed that the new peak shear strength criterion has the capability to predict anisotropic peak shear strength accurately. In practice, to allow for modeling uncertainties, the new criterion should be used with a factor of safety of about 1.5. |
