C.F. Lee

Displacement back analysis for a steep slope at the Three Gorges Project site

Abstract This paper contributes to two aspects of displacement back analysis of rock slopes. First, a novel method for displacement back analysis is presented, based on error back-propagation neural network and genetic algorithm (GA). The BP network replaces the time-consuming finite element method, thus enhancing the efficiency of the analysis. The GA is used as an optimization method, whose global search strategy can effectively improve the reliability of the analysis, thereby, making the back-analyzed results independent of the initial values. Application of this methodology is illustrated with a numerical example. Secondly, the deformation mechanism of rock mass is duly taken into account in the displacement back analysis of Profile 17–17 of the permanent shiplock slope at the Three Gorges Project site, by treating the disturbed zones as weakened media, yielding reasonable results.

Estimating the excavation disturbed zone in the permanent shiplock slopes of the Three Gorges Project, China

Abstract The stability and deformation of the permanent shiplock slopes are among the key issues in the design and construction of the Three Gorges Project. The permanent shiplock slopes, formed by deeply and steeply cutting into weathered and fresh granites, are 1607xa0m in length and 50–170xa0m in height. The disturbed zone in the permanent cut slopes induced by excavation is therefore one of the most important aspects in slope stability and deformation evaluation. A comprehensive investigation including non-linear finite element analysis, in-situ testing, instrumentation and monitoring, back analysis and rock mass quality rating has been carried out for the identification and evaluation of the excavation disturbed zone in the permanent cut slopes. The results of the investigation confirmed the existence of such an excavation disturbed zone in the permanent cut slopes. This zone is characterized by a considerable weakening in the mechanical properties of the rock mass. From the cut surface to the deeper region of the permanent cut slopes, the excavation disturbed zone can be further divided into a damaged zone, an affected zone and a slightly affected zone according to the extent of weakening in the rock mass. The damaged zone and the affected zone have thicknesses of 5–10xa0m and 10–20xa0m, respectively. The exact pattern of the excavation disturbed zone is variable in different parts of the permanent cut slopes and are generally similar to that of the plastic zones estimated by using the non-linear finite element analysis. The present investigation has provided both factual data and insights for the stability and deformation evaluation of the permanent shiplock slopes. The approach and methodology developed in the paper can be used to assess similar excavation disturbed zones in other large cut slopes.

Quantifying topography and closure deformation of rock joints

Abstract This paper presents a study for quantifying both the joint topography characteristics and the load–closure deformation of a rock joint under normal compressive loading condition. The study covers (1) laboratory measurements of rock joint surface profiles using a profilometer designed and fabricated by the research team, (2) development of a mathematical method to identify the waviness and unevenness components in joint surface profiles and the associated composite topography, (3) development of a general load–closure deformation model by using both the waviness and unevenness components in the composite topography, (4) unconfined compressive testing of rock samples with joints for the experimental load–closure deformation of joints and, (5) verification of the general load–closure deformation model by the experimental load–closure deformation results. The study leads to the following four findings: (a) the mathematical method can be used to identify the waviness and unevenness components for joint surface profiles and its composite topography. The height characteristic parameters of the complete surface topography of a joint are mainly determined by the waviness component. The texture characteristic parameters of the complete surface topography of a joint are mainly determined by the unevenness component, (b) joints can be classified into the three contact state cases using the waviness and unevenness components for both the joint surface profiles and the associated composite topography. The load–closure deformation behavior of a joint is determined by the waviness and unevenness components of the composite topography for a specific contact state, (c) the general load–closure deformation model developed in this paper is applicable to the three contact state cases. The general load–closure deformation model uses the composite topography of a joint and can take into account the effects of the contact states, the initial aperture and the waviness and unevenness components; (d) parametric studies and verifications with the uniaxial compression test results show that the general load–closure deformation model gives reasonable estimations of the load–closure deformation behavior of rock joint under compressive loading and can be reduced to those given by other researchers in the relevant literature.

Mechanical behavior and microstructural variation of loess under dynamic compaction

The laboratory dynamic compaction test was used to study the effects of dynamic impact loads on the shear strength characteristics of loess. Loess specimens were compacted by different numbers of blows and their shear strengths measured. The experimental results revealed that the shear strength varies in stages with the number of blows. During the initial stage, the shear strength increases with the number of blows. It reaches a peak value and then decreases with the number of blows. To reveal the relationship between this alternation in shear strength and microstructural variation, quantitative microstructural analyses were performed on some tested specimens at different stages of strength evolution. This microstructural study demonstrated that the variation of shear strength is directly related to the alteration of microstructure. It is concluded that the shear strength of loess is basically controlled by its microstructural state. Growing particle size heterogeneity, varying particle orientations and increasing microstructural damage are the main causes of the alternation of shear strength.

Three-dimensional back-analysis of displacements in exploration adits — principles and application

Abstract Back-analysis from displacements has been attracting the interest of the rock engineering profession in recent years. It has been applied since its inception in the 1970’s because of its instinctive practical value to the designer. In this paper we discuss the principles and application of one feasible method, that of Three-dimensional displacement Back-Analysis (TBA), which is used to back analyze the modulus of elasticity ( E ) of the rock mass and the horizontal component ( P ) of rock stress mainly by means of displacements measured during the the excavation of an exploratory geological adit. An attempt is also being made to solve the problem of unique resolution of TBA.

Deformation of the central pier of the permanent shiplock, Three Gorges Project, China: an analysis case study

The central pier (CP) is one of the key structures of the permanent shiplock in the Three Gorges Project on the Yangtze River, China. This paper deals with the study of the rock mass deformation of the CP of the second shiplock under various operation conditions. Numerical simulations using the 2-D distinct-element method and 3-D finite-element method are applied to analyze the rock mass deformation. Also, time-series analysis and gray system theory are used to predict the trends of rock displacements based on in situ measured data. The results can be summarized as follows: (1) the main deformation of the rock mass in the CP occurs after excavation completion, while the shiplock is in operation, the dominant displacements mainly occurring on both sides of the northern chamber, developing generally towards the chamber; (2) the horizontal displacement varies with the process of excavation, and is in a steady state after excavation, and during operation of the shiplock the displacement is slightly decreased; (3) the horizontal displacement parallel to the axial line of the shiplock occurs downstream; and (4) the vertical displacement occurs during excavation, and then subsidence with various values takes place in different parts of the CP for different operation conditions.

Assessment of an equivalent porous medium for coupled stress and fluid flow in fractured rock

Abstract The equivalent porous medium (EPM) for coupled stress and fluid flow in fractured rock was assessed. The assessment was focused on the contributions of the seepage-induced force to the systems equilibrium and deformation as well as fluid flux through the fractures, with the special emphasis on the inter-relations among these three aspects. Seepage-induced average stress was calculated for the EPM. And it was proved, with respect to the fluid flow system (FFS) proposed by Long et al. [Long JCS, Remer JC, Wilson CR, Witherspoon PA. Porous medium equivalents for networks of discontinuous fractures. Water Resour Res 1982;18:645–658], that the equivalence of the equilibrium contribution of the seepage-induced force is unconditionally satisfied in the EPM. The equivalence of fluid flux and seepage-induced deformation was evaluated using numerical techniques based on the FFS, in which the seepage-induced deformation was considered as ‘tensile’ deformation under fluid pressures. The numerical simulation results of the two cases suggest that the EPM established in terms of fluid flux based on the FFS guarantees equivalence of coupled stress and fluid flow. Also, the numerical simulation results indicate that the percolation theory can not be applied to a system with fractures of remarkably different size and hydraulic conductivity.

Spatial distribution of inter-layer shear zones at Gaobazhou dam site, Qingjiang River, China

Abstract The spatial distribution of inter-layer shear zones is studied, based on the fully mapped inter-layer shear zones at the Gaobazhou dam site on the Qingjiang River in Hubei province, China. The field observations revealed that two types of inter-layer shear zone, under-developed and well-developed, are distributed differently. The former are associated with the incompetent rock layers, and the latter are controlled by the inter-bed folds and lithological properties of the strata. To represent their lithological properties geomechanically, data on strength by the Schmidt hammer test and thickness of each field-identified rock layer were collected to form strength sequences and thickness sequences. The adjacent rock layers of similar strength are then grouped, and two grouped sequences are combined into a single sequence, whose individual term is the strength weighted by the exponential function value of the centralized thickness (SCT). The rescaled range ( R / S ) analysis results of SCT (SCT-R/S), taking into account the presence of inter-bed folds, are in good agreement with R / S analysis results of the spacing sequence of the well-developed inter-layer shear zones. Thus the SCT-R/S model has the potential to predict their spatial distributions.

Particle-size distribution of interlayer shear zone material and its implications in geological processes—a case study in China

Abstract A case study was carried out at the Gaobazhou dam site on the Qingjiang River in Hubei Province, Central China, to investigate the particle-size distribution (PSD) of interlayer shear zone material and its implications in the geological processes. Forty-six disturbed samples were taken at various representative locations within the well-characterized interlayer shear zones at the Gaobazhou dam site. The particle-size distributions of these samples were analyzed, and the mass-based approach is used for fractal scaling of the particle-size distribution curves. It is shown that all curves on the log–log scale can be approximated by two intersecting lines. Thus, each curve can be quantified by α 1 , α 2 , P s , and r s , which respectively correspond to the slopes of the two lines and the coordinates of the point of intersection. α 1 and α 2 can be replaced by fractal dimensions D 1 and D 2 . The statistical analysis results reveal that Delta (= D 2 − D 1 ) and P s are closely related to D 1 . Therefore, only D 1 and r s are used to characterize each curve. It is illustrated that 47 specimens are clustered into five groups with respect to r s and D 1 . Accordingly, five categories of particle-size distributions are identified. It is found that the D 1 and r s values of a sample can reflect the geological processes of the interlayer shear zone. The larger the value of D 1 , the higher the degree of interlayer shear experienced. The value of r s implies the degree of subsequent secondary geological processes, but in a rather complex way due to the differences in intensity of interlayer shear undergone. Based on this, the geological evolution processes of the interlayer shear zones are postulated, and the current status can be defined through fractal scaling of the particle-size distribution of disturbed samples.

Finite Element Analysis of Landslide in Dredged Slope

This paper presents a finite element analysis of a landslide that occurred in a newly dredged submarine slope for port development. The landslide was a multiple retrogressive landslide. The marine slope comprised recent marine and river mouth delta deposits. The landslide occupied a plan area 200 m long and 150 m wide. A volume of about 240,000 m3 soil slipped into a newly dredged open space in the sea. The finite element analysis indicates that the backfill pre-loading significantly increased the failure zone in the dredged slope and the soil lateral displacements toward to the open space in the sea.