| Seismic-induced layered rock slope failure is a common type of geo-hazard worldwide, it can cause severe damage for human ware fare. Research on its failure mechanism involves several different scientific topics and has been one of the most complex problems in the geotechnical research field. Up till now the methods used in the seismic-induced layered rock slope failure mechanism are borrowed from the research on soil slope dynamic failure, these methods, however, cannot properly represent either the geological characteristic, or the failure process of rock masses with rock layers and joints in them. To concur this shortage above, this thesis focus on explaining the dynamic failure mechanism of layered rock slope with the respect of the natural rock mass structuralcharacteristic, by utilizing the theory of geotechnical analysis, rock mechanics, rock fracture mechanics, as well as physical and numerical model tests, to study the seismic-induced failure mechanism of three basic type of layered rock slope:bedding slope, inverse slope and level slope.Based on the theory of rock-joint structural control, analyzed the basic joint structure network of the three type of layered rock slope, and divided the rock joints in the layered rock slope into two predominant kind:rock plane and secondary rock joint. The research showed that these two predominant rock joint types are basically vertically lied to each other, and both two types consist of unattached part and attached part inside them. The study on the effect of the secondary rock joint shows that this kind of joint plays an very important role on the control of the rock slope safety. And the study on the effect of the rock plane shows that the attached part inside the rock planes serve to remarkably increase the safety of the rock slope.Based on the theory of rock mechanics and rock fracture mechanics, the failure process of rock joint is deduced, and proved by some rock fracture tests carried by former researchers. The theoretical results show that rock fracture can only produce Mode I tensile failure, and the so-called Mode II’shear’failure proposed and claimed by some researchers is essentially the combination and connection of numerous small Mode I tensile fracture. The dimension of the so-called’mode II rock fracture’is beyond the dimension of the concept of classic fracture mechanics, therefore is not real’shear fracture’. This difference reveals that the rock fracture mechanics should be viewed as a certain type of material scientific theory whose study dimension isbetweenmicroscopic and macroscopic. The fracture processing formula and failing judgment are developed for the microscopic rock plane fracture under different type of stress conditions and with different strength compared with the intact rock, and its processing form is carefully studied and explained by rock fracture mechanics. The results show that the strength of the non-through rock plane is essentially controlled by the stress conditions of the rock plane; and the rock plane fracture failure type can be affected by the ratio of the strength of the rock plane and the strength of the intact rock. Improved the structural stress compute model for the formation of secondary rock joint, and studied the fracture process of the secondary rock joint under different structural stress condition. By utilizing the theory of rock fracture mechanics, found two explanations of why secondary rock joint usually cannot cut through multiple rock layers, these are:1) secondary rock joint cannot cut through the already unattached open rock plane; and2) when the secondary rock joint progress and encounter the attached rock plane, the secondary rock joint will continue to progress along the rock plane in spite of the stress condition it is behold, this is due to the fact that in the reality, the strength of the attached rock plane is far smaller than the strength of the intact rock.This thesis identified the characteristics of the dynamic failure of the bedding slope, inverse slope, and level slope, and improved the basic compute model of these three type of rock slope. Based on the theory of rock-joint structural control, classified and analyzed the rock plane and secondary rock joint dynamic failure mechanism during the process of the seismic-induced three basic type of layered rock slope failure.Two basic type of layered rock slope physical model-the sliding and inverse slope with rock plane and secondary rock joint in them-were built of synthetic material, and tested in the centrifuge testing system to witness the failure process of the rock slope during earthquake. Basic rock sample tests and fracture progressing tests to the synthetic rock mass material proved that the material recipe, the casting procedure, and the newly invented method to build closed rock plane and secondary rock joint are suitable for presenting the important characteristics of real layered rock mass. The synthetic material is made of gypsum, sand and pure water with certain mix ratio through straight casting procedure, its physical and mechanical properties are similar with sandstone; New methods are invented to make the rock plane with no gap; the attached parts inside the rock plane are made with good position control and strength control. several advances are made for the centrifuge testing system to enable the dynamic test for rock like material and rock slope model:a brand new model loading plate was designed and built; and a new monitoring system was designed and set up to acquire the accurate breakage time of the rock joints inside the physical model during the test. The results of the centrifuge dynamic tests reveal that the topographic amplification effect of the layered rock slope is related to the frequency and amplitude of the input seismic motion, the reason for these relations can be united to the damping function of the slope on the topographicamplification effect:the damping function of the slope is related to the input motion, and as the damping function grows bigger, the amplification effect becomes more obvious. Moreover, the test results show that the secondary rock joint inside the rock slope physical model plays a very important role on both the dynamic response and the dynamic failure process of the two different rock slope type, it serves to decrease the dynamic stability of the two type of the rock slope.The hybrid finite-discrete model simulation method is provided and improved to simulate the seismic-induced layered rock slope failure process, in which PFC2D is used as discrete element simulation software and FLAC is used as finite element simulation software. The relationship between the property of the PFC2D intact rock model(which is the union of numerous particles) and the property of the PFC2D particles is carefully studied, and the Smooth Joint Contact Model of the PFC2D is improved in order to add attached and unattached rock plane and secondary rock joint into the PFC2D intact rock model, this improved model is tested by carrying out several group of certain numerical test and comparing the test results with the theoretical results deduced in this thesis, and the result shows close agreement between them.The hybrid finite-discrete element models for bedding slope, inverse slope and level slope are built and tested under horizontal seismic load, the failure process and the effect of rock plane properties and secondary rock joint properties to the dynamic failure process and safety of these three type of layered rock slope are carefully studied, the tests results are as below:During the dynamic failure process, the bedding slope mainly had sliding movement along the failure plane which is the combination of the opened rock planes and secondary rock joints. Some of the attached rock planes failed in shear, but there are also ineligible number of attached rock planes failed in tension; the secondary rock joints mainly failed in tension, rarely in shear. The strength of the attached rock plane and the continuity rate of rock plane effect the dynamic stability of the sliding rock slope significantly, while the shear strength of the unattached rock plane barely has any effect; the strength of the attached secondary rock joint and the spacing of the secondary rock joint effect the slope’s dynamic behavior significantly, while the shear strength of the unattached secondary rock joint serves very small effect. Tests results shows that a large number of rock planes inside the bedding rock slope failed in tension even under the horizontal seismic load used by this thesis, which failed to be revealed in the original failure mechanism theory of bedding slope failing in pure sliding along the rock planes. Therefore, the tensile strength should be treated as one of the controlling factors to the dynamic stability of bedding rock slope under seismic load. The failure process of the bedding slope during the simulation is a gradual process, as the seismic input motion ramped, the failing zone of the slope extended from the surface of the rock slope gradually to the inner of the rock slope, and multiple cut-through failure plane formed during the failure process, due to the internal breakage of the rock masses. Therefore, the classic slope stability analysis method cannot cover the actual dynamic stability of rock slope, because it usually consider the rock slope fails along one single potential failure plane, and calculate the slope stability with this pre-specified failure mode. In order to solve this problem, an amended rock slope sliding failure criterion is invented and tested by several simulation results. This amended failure criterion include two valuation parameters:1) the critical amplitude of the seismic input motion for the rock slope to form the first cut-through failure plane; and2) the volume of the failing rock mass cut by this failure plane. By using this criterion, both the dynamic stability and the predicted destructive range of the rock slope can be judge properly.During the dynamic failure process, the inverse slope mainly had toppling failure along the step-path failure plane at the bottom of the rock slope. The rock plane mainly had shear failure during the failure process, with less amount of tensile failure also, and most of the tensile failure of the rock plane occurs in the crest part of the rock slope on top. The rock mass on the top of the rock slope generally broke along the secondary rock joints, to form numerous tensile cracks, and the whole rock mass on top toppled during the failure process, generated large rotation; the secondary rock joints at the bottom part of the rock slope generated both shear and tensile cracks, and had obvious slipping movement but with very small rotation. The strength of the attached rock plane and the continuity rate of rock plane effect the dynamic stability of the toppling rock slope significantly, while the shear strength of the unattached rock plane barely has any effect; the strength of the attached secondary rock joints, the shear strength of the unattached secondary rock joints and the spacing of the secondary rock joints all had relatively small effect the slope’s dynamic behavior. During the dynamic test, the rock mass nearly the crest of the slope first generated secondary joint tensile failure, which cut the rock layers through and made connections with unattached rock planes, made the rock mass have the tendency to topple; as the dynamic input motion ramped, the secondary joints inside the bottom of the rock slope started to fail in shear or tension, and a cut-through failure plane beneath the inverse rock layers was formed gradually, enable the rock layers at the bottom to slide along this failure plane, and triggered the total toppling failure of the whole slope. This failure process is basically inversed to the failure process of inverse rock slope under static load, which by static analysis shows that the inverse rock slope usually failure firstly at the bottom, inducing support loss for the upper layers, and then the upper layers bend and topple gradually. This difference above reveals two major facts for the dynamic failure mechanism of inverse rock slope:1) the presence of the secondary rock joints can severely affect the failure mechanism of inverse rock slope; and2) the static failure process and the dynamic failure process of inverse rock slope are very different from each other, cannot be treated as the same.During the dynamic failure process, level slope generated many internal breakages inside the rock mass, which include both shear and tensile failure of rock joints, and the tensile failure is superior in number. The rock mass first formed a large number of open cracks whose orientation are nearly vertical, and as those open cracks grows longer and the number grows bigger, they can connect with each other to form a macroscopic’shear’failure plane inside the rock slope, which eventually cut through the rock slope in a shape of circle. The dynamic stability of the level slope is controlled mainly by the strength of the secondary rock joints, and the size of the failing rock mass is strongly affected by the thickness of the weak rock mass on the surface of the level slope, when the thickness is bigger, the size of the failing rock mass during when dynamic failure occurs grows larger. To the contrary, the shear strength of the rock planes barely has any effect on the stability of the level slope. However, the dip angle of the level rock slope serves to effect the permanent displacement of the slope during the dynamic failure process, as the dip angle changes the level slope from slice-bedding to flat-inverse, the permanent displacement of the level slope under the same seismic load at the same time decreases in negative exponent form. Therefore, level rock slopes with a relatively small change of the dip angle of the layered rock masses could have a very different critical permanent displacement when the slopes are on the threshold of failing, and it is impossible to find one unique critical value for all level slopes to be used to judge their dynamic stability.Finally, one example simulation work is worked out to illustrate the properness of the combined finite-discrete element method on simulating layered rock slope dynamic failure process. The Sunjiayuan level rock slope failure during the Wenchuan Earthquake is picked as this demonstration work, and the horizontal Wenchuan Earthquake record is used as the input motion for this test. The failure process of the Sunjiayuan level rock slope numerical model shows that during the earthquake, the upper part of the rock slope failed of sliding along the circle-like failure plane, and rush down along the mountain, shoveling weak rock masses on the surface of the mountain, blocked the river downside, then stopped and deposited along the valley. The test result shows good consistency with the actual event. |
PDF Full Text Request