In this study the concept of equivalent second order asperities was devised and incorporated to the elasto-plastic joint model. Two asperity angles and two damage coefficients were introduced to handle the degradation of two types of asperities. Based on this formulation monotonic and cyclic shear behaviors were numerically simulated and compared with the test results of rough rock joint specimens. Characteristic shear behaviors of rock joint could be well simulated by utilizing the degradation of first and second order asperities. Variations of joint permeability with the shear displacement were also investigated using the hydraulic model which could consider the gouge production and aperture change.
A number of joint models, which can consider various features of joint characteristics, have been developed to analyze the behavior of discontinuous rock mass. After Patton (1966) proposed bilinear models based on saw-tooth type joints, several empirical and theoretical constitutive models (Ladanyi & Archambault 1970, Barton 1977, Piesha 1987,
Jing et al. 1993, Qiu et al. 1993) have been reported. Barton's empirical model has been broadly used, because it includes several important factors for joint characteristics. However, since Barton's model is not formulated incrementally it is somewhat difficult to implement numerically. Plesha's model (1987) idealized Patron's saw tooth type asperities and was based on the classical plastic theory. In his model, the asperity degradation during direct sheafing is an exponential function of the exerted work. This model has many advantages such as theoretical completeness, reasonable realization of joint behavior and adaptability to numerical implementation. However this model over-predicts the dilation during cyclic shearing, and has not been evaluated for the behavior of rough rock joint. There are few coupled hydro-mechanical model compared to mechanical one. Barton's dilation-conductivity model (1985) over-predicts conductivity changes.
In this study Plesha's elasto-plasfic joint model was modified to incorporate the concept of equivalent second order asperities. Laboratory monotonic and cyclic shear behaviors were simulated with the proposed model. Variations of joint permeability with the shear displacement were also investigated using the hydraulic model, suggested by Nguyen & Selvadurai (1998), which could consider the gouge production and aperture change.