This paper presents the results of the coefficient of restitution and the kinetic energy loss rate obtained by lab experiment, two parameters that are crucial for rockfall impact. However, various definitions of coefficient of restitution exist and the most appropriate one is still not formed and obtained. In addition, the energy variation during the rockfall impacts has important significance in practical design. In this research, two kind shapes of blocks including plate and strip were adopted in the laboratory testing and the block material was tested before, indicating that the material has sufficient strength to prevent shattering during the impact. Furthermore, an apparatus specifically built for this study was established including a base, a slope and a releasing device. The falling testing was performed using plate and strip block while the falling height as well as the slope angle and releasing height were altered during the tests in order to estimate the effect of each parameter on the coefficients of restitution and energy loss rate. It was observed that collision reflection angle is less than impact angle for all, suggesting energy loss in collision. Impact angle decreases with increasing slope angle while there was no obvious effect of releasing height and releasing angle on impact angle. The relevant coefficient of restitution was found to decrease with impact angle, and the kinetic energy loss rate increased. Finally, the kinetic energy before and after the impact was found to significantly affect the COR and energy loss rate and the results can provide basis for mitigation measures.

The discontinuous joints are an essential type of natural joints. The normal force, joint persistency, and distribution exert great influences on the shear resistance of the rock joints. By simulating the uniaxial compression experiment and Brazilian test, the material parameters and the basic size standard for meshing were determined. The symmetrical discontinuous joint distribution of three types were established, the cohesive elements were inserted between the solid elements, and the numerical simulation of the shear test was conducted. The effects of joint distribution, joint continuity, and normal stress on the shear resistance of joint rock were investigated, and the law of crack evolution was analyzed. The results showed that the shear process of discontinuous joints can be divided into four stages: elastic stage, strengthening stage, plastic stage, and residual stress stage. For the scattered joint distribution, the rock bridge can provide more reinforcement for the joints, which enhances the shear resistance of the joints, the stress concentration point at the end of the joint is easy to accumulate more fracture energy, which induces the initiation of the cracks, and under the influence of unbalanced torque, the both-sided joint distribution is more likely to produce tension damage.

Deformation of surrounding rock is widely monitored to discover surrounding rock behaviors for purpose of event forecasting. This article aims to present a comparative study on surrounding rock nonlinear deformation prediction using computational intelligence techniques. The Gaussian process (GP), the support vector machine (SVM), and the wavelet neural network (WNN), are analyzed comparatively for predicting the surrounding rock deformation series. Two representative tunnels, the Wangdeng tunnel on Chenglan railway in China and the Ureshino tunnel line I on Nagasaki expressway in Japan, are illustrated. The results prove that the computational intelligence approaches are capable of predicting surrounding rock nonlinear deformation. The GP, on the whole, performs best. The SVM shows better ability than the WNN and GM (1, 1) not only in predicting the settlement and convergence deformation values but also in tracking the trends of surrounding rock deformation curves.