IACGE 2013: Challenges and Recent Advances in Geotechnical and Seismic Research and Practices

Aiming at some disputation about disposal methods on vertical draining channels in model of numerical analysis of vacuum preloading, some analysis was carried out in the paper. Through a practical key project, two different disposal methods of vertical draining channels were used in the project. The finite element software was used to simulate the consolidation process of soft-soil in site. The results of simulating calculation and results measuring in site were compared. Differences of the two methods were shown clearly in the paper. Applicable instances for the two different disposal methods were pointed out.

Dynamic responses of soil foundation are difficult to analyze, though it is related to kinds of practices. In this paper, dynamic responses of soil foundation under dynamic load were analyzed theoretically. Perturbation method was used to obtain the theoretical solution. Controlling equations were about liquid-solid two-phase media. Flow, function, and potential function were introduced to decouple the controlling equations. Then, perturbation expansion was introduced into the equations of flow, function, and potential function. The responses' characteristics are discussed.

Deposited sediment shows significant large strain characteristics during weight consolidation with variable consolidation parameters. Non-neglected prediction errors will appear if applied classic Terzaghi theory in large strain consolidation calculation, especially as the settlement amount exceeds 50% which will lead to great coordinate errors and can't reflect the nonlinear features. In this paper, a new piece-consolidation model (based on modified Terzaghi theory) is proposed for deposited sediment consolidation prediction, of which large strain consolidation process is divided into a series of small strain consolidation, and each strain step consolidation is predicted by Terzaghi theory with necessary parameters modification. Power functional equation is applied for the constitutive relationship between permeability, compressibility and void ratio. With the assumption that each strain step consolidation coefficient is constant, large strain consolidation process can be determined by all the strain step settlement accumulation. Piece-consolidation model based on Terzaghi theory successfully meets the basic assumptions and reflects the nonlinear features. Moreover, model outputs also show a well agreement to the experimental measured.

A rotating block method is developed to calculate the rotational displacement of slopes of saturated clay under seismic loading. The method is similar to the pseudo-static sliding block method of Newmark and is quite easy to use in design calculations. When the threshold acceleration for rotation is exceeded, the slope will start to rotate until the angular velocity for rotation is reduced to zero. Rotational displacement will be accumulated each time the threshold acceleration is exceeded during ground vibration. The influence of ground motion characteristics on the computed slope deformation is evaluated. Implications for design are discussed.

The attenuation plateau formed during radiation of dynamic compaction shockwave has a severe impact on construction work in loess backfill site and safety of buildings nearby. With essential targets of finding out reasons for attenuation plateau formation and providing basis for safe distance determination, geophysical prospecting was adopted in the test for in-depth investigation on the attenuation plateau characteristics of shockwave so as to provide basic data for attenuation law study of shockwave. Appropriate cut-off frequency was selected according to the test results and energy composition characteristics of elastic wave in semi-infinite soil. The shockwave energy was divided into body wave energy and surface wave energy in order to analyze energy attenuation characteristics of different waves. It is confirmed that the main reasons for attenuation plateau formation in the area were proportion and radiation characteristics of body wave energy. On this basis, staircase method was used to fit the radiation attenuation characteristics of dynamic compaction shockwave through subsection regression and compare with traditional fitting methods. The results showed that in close-distance and high-density dynamic compaction work, staircase method can be more accurate to mark out safe distance.

The resilient modulus and Poisson's ratio of subgrade soils in highway foundation are important parameters in design and quality control process. The currently used techniques include California Bearing Ratio (CBR) test, resilient modulus test, Dynamic Cone Penetrometer (DCP), and Falling Weight Deflectometer (FWD) tests. However, these techniques have certain limitations and sometimes fail to satisfy the requirement and accuracy for design purposes. In recent years, piezoelectric sensors have been widely used for laboratory measurement of wave velocities in soil and rock specimens. For shear wave velocity determination, bender elements have been applied to the tests. On the other hand, extender elements have been used for the measurement of primary wave velocity. This new laboratory testing technique is adopted to measure the two important parameters, the resilient modulus and Poisson's ratio of subgrade soils in laboratory tests. The results of using this technique on a soil sample is presented and compared with that obtained from CBR tests. It is concluded that this new technique is simple, accurate and has potential to be widely used in engineering practice.

In this paper, a reasonable train dynamic loading form is determined, and a finite element model of PCC pile composite foundation for high speed railway is established using ABAQUS software. The dynamic response is simulated on subgrade and pile-soil composite foundation under different train speeds. The dynamic stress and acceleration changes of composite foundation are analyzed by changing the modulus of the roadbed surface layer and cushion. The results show that the changes of train speed, stiffness of roadbed and cushion layer react differently according to the laws of distribution and dispersion for dynamic stress, acceleration response of subgrade and pile-soil composite foundation. The dynamic stress and acceleration in roadbed surface increase with the increase of train speed. The changes in stiffness of roadbed have little effect on dynamic stress of the lower pile area, while the cushion is significantly influenced by as much as 90%. The cushion has the effect on adjusting pile and soil dynamic stress ratio which may result in up to a 50% change in difference. A dynamic stress arch effect appears in the cushion under the dynamic interaction of the pile and cushion layer, which is similar to the static soil pressure arch effect.

In this paper, a three-dimensional dynamic coupling finite element model of tracks of embankment pile-soil composite foundation has been established. The dynamic displacement of pile composite foundation is analyzed based on modulus parameter changes of foundation bed, pile, cushion layer and foundation. The results show that the interface area between cushion and pile increases with the decreasing of pipe pile top, the load-sharing ratio of the PCC pile increases significantly, and the load that the soil bears also decreases. The magnitude and scope of dynamic displacement for subgrade also decreases. It is worth noting that the dynamic interaction between PCC piles and cushion layer is greater than that between soils and the cushion layer. Rigid piles tend to penetrate into the cushion layer; the dynamic curve along the cushion layer surface is similar to type 'W'. An increase in modulus of the cushion enhances the dynamic interaction effects between PCC pile and the cushion, which, to some degree, encourages PCC piles to share more bearing responsibilities. This mechanism magnifies the dynamic displacement of pile top and body, whilst simultaneously reducing upward penetration deformations.

Since May 12^th, 2008, it has been already determined that tens of thousands of landslides were triggered by Wenchuan earthquake. However, full understanding of the mechanism of seismic landslides is still unclear due to the complex geological condition and dynamic action of seismic wave. This paper presents the Liujiawan landslide accumulation characteristic and intends to analyze the formation mechanism under the Wenchuan earthquake. Based on field investigation, a unique geological condition of Liujiawan landslide appears two stage platform accumulations with different lithology, which have an obvious landslide boundary in lithological interface and unloading weathering zone. These characteristics are hardly found in other types of landslide. In the meanwhile, a numerical model displays tension stress concentration on the lithology boundary surface. From the results, it can be concluded that the seismic waves propagate through the different physical properties of interface and structure plane, which mutated strong dynamic stress and induced slope failure. The dynamic stress mutations should be considered as the triggering factor of Liujiawan slope failure.

Based on 346 complete data sets collected from 27 loess tunnels along the Zhengzhou-Xi'an Passenger Dedicated Railway, a new method for classification of loess ground is proposed by using multidimensional scaling (MDS) and principal component analysis (PCA). The comparative studies are conducted by the use of either only 7 tested indices or all 12 indices (i.e. 7 tested indices plus 5 deduced indices). Both the MDS and PCA provide similar but reasonable explanation of links between different indices. The first principal component (F1 or FPC) is adopted as a classification index to separate the total data into 5 groups according to the frequency distribution of F1 by using the golden section concept. Better and more reasonable results can be obtained from the classification by F1 from 7-index PCA than those by 12-index PCA, which not only verified both the importance of independence among indices to classification results and cohesion to loess characterization, but also revealed the clear boundaries of different classifications of loess ground from the statistical perspective. This study indicates that the cohesion should be considered as an optimum index to reflect the intrinsic strength characteristic of loess.

Over the years, a number of researches have been carried out to describe the behavior of granular soils. However, the mechanical behavior of cemented granular soils is not well understood. Indeed, shear strength and dilation characteristics are different from uncemented soils. The lack of experimental data for illustration of the mechanical behavior of cemented soils is more obvious for cemented sand-gravel mixtures.In this paper, shear strength-dilation behavior of artificially cemented sand-gravel mixtures has been studied using large direct shear tests. The soil consists of fine sand mixed with 30% uni-sized gravel particles with maximum size of 12.5 mm. Large direct shear tests have been conducted using saturated samples in a 300×300×170 mm shear box. Samples were prepared in four different relative densities of 30, 50, 70 and 90% and three cement contents of 0, 1 and 2% and were tested under two surcharge pressures of 77 and 150 kPa. Cemented samples were cured in direct shear box for seven days under a low water head to prevent degradation of cemented bonds. After curing, shear loading was applied at a constant rate of 0.1 mm/min. Both horizontal and vertical displacements besides shear stresses were recorded during loading process. Shear strength, friction angle and cohesion intercept of cemented samples were calculated and it was indicated that shear strength of cemented sand-gravel mixtures increase with cement content and relative density. However, dilation of cemented samples reduced with increase in surcharge pressure.

It is widely believed that differential settlement at railway transition zones is the result of dynamic wheel/rail force, and the stiffness difference between embankment and fixed structures is believed to be one of the major causes of high wheel/rail forces. A dynamic simulation model, which takes into account both a varying subgrade stiffness and track settlement, and incorporates the dynamic characteristics of railway vehicles, was applied to analyze the influences of subgrade stiffness on dynamic loads at railway transition zones. The studies reveal that stiffness variation at transitions is of minor influence over dynamic wheel/rail forces, and the overall dynamic effect is dominated by track settlement differences at transitions and the stiffness of fixed structures.

In order to study the characteristics of dynamic stress in subgrade bed, a large-scale, ballastless track dynamic model test was made based on the Suining-Chongqing high-speed railway. The results indicate that the dynamic stress has an uneven distribution along subgrade cross-section, and the unevenness is more and more obvious as the dynamic loading increases. Along the depth direction, the dynamic stress in the surface layer of subgrade bed decays more rapidly than in the base layer. The dynamic stress can be calculated by Odemark and Elasticity theory as its distribution under foundation slab of ballastless track is simplified uniform along lateral direction, and triangular along longitudinal direction. On the premise of dynamic stress was known, this paper assumed each soil layer of subgrade to be one-dimensional compression model and determined the limited thickness of compression layer. Based on this, an approach of conversion from multi-layer system to the equivalent Winkler foundation was explored, and the coefficient of subgrade reaction in different working conditions can be acquired.

Test soils were collected from a debris flow source area in the Weijia groove, Beichuan County, Sichuan Province. An indoor artificial rainfall model tank test has been designed; it can be used to simulate the process of debris flow initiation for slope loose soil under the rainfall action. The debris flow starting condition and confluence process have been measured in this test through the analysis of the relationship of the rainfall intensity, duration time before the starting, and the total rainfall capacity at the starting moment, the relationship between the volume of production flow and the wate-to-soil ratio in production flow. All these research results can be used to simulate debris flow initiation mechanism obtained from model test focused on the seepage-induced action and its influence factors.

Numerical modeling and Response Displacement Method are applied to assess the effects of the presence of cross passages and geology formations on the seismic responses of the Qianjiang River tunnel, in order to determine the proper locations of cross passages. The numerical analysis shows that the presence of cross passages affects the longitudinal seismic response of the main tunnel significantly within a 25-m distance from the intersections of cross passages with main tunnels. So, the possible local failure of the intersections will be a major concern.This is largely attributed to the complicated geologic formations and irregular ground conditions. The analysis based on the Response Displacement Method indicates that cross passages in flat and uniform ground layers can mitigate the local failure at the intersections. Optimal location for cross passages are proposed based on the detailed analysis of the geologic profiles of Qianjiang River Tunnel.

Based on the discrete element numerical simulation, the change rules of safety factor and failure modes of slope influenced by parameters of rock mass structural plane are studied. Based on the principle of variance analysis of orthogonal experiment, significance of the rock structural plane parameters to the stability of slope is studied. It is shown that slope safety factor is linear increase in a certain range with the increase of the strength of the rock mass structural plane, and the failure modes shift gradually from the bedding sliding failure modes to the sliding-bending failure modes. The extent of variation of safety factors changes very little with the increase of the normal and shear stiffness and spacing of rock structural plane. The slope failure modes are mainly sliding failure modes. Slope safety factor first decreases and then increases, and finally decreases with the increase of the rock structural plane dip angle. Failure modes shift from shearing slip failure modes to shearing slip and buckling failure modes, and finally to the tilting failure modes. The impact of the rock structural plane cohesion to the slope stability is the greatest, and the stiffness is the least.

The geocell flexible retaining wall has a broad application prospect in the road slope protection. However, its failure mechanism is not systematically investigated. In this paper, by employing the FEM software Plaxis, the failure mode of the geocell flexible retaining wall is studied by the numerical analysis. The effects of soil strength, geocell structure strength, and location of high strength geocell structure layer on the failure mode are investigated based on the analysis of the numerical results. The research results indicate that the sliding surface is a curved surface which passes through the wall toe and intersects with the wall, which differs from the planar one in the Coulomb's soil wedge theory. Both soil strength and geocell structure strength have great effects on the location where the sliding surface intersects with the top of the backfill surface. However, their effects on the location where the sliding surface intersects with the wall back are not significant. In addition, the failure mode of the wall with the lower half part constructed by means of high strength geocell is almost the same with that of the whole wall constructed with high strength geocell. The results in the study will be useful for the determination of the sliding surface of the geocell flexible retaining wall when evaluating the earth pressures against it.

Dynamical systems-based soil shear holds that rates of change of shear stress, effective normal stress of a sheared mass of particles are proportional to applied values of shear and effective normal stress; initial proportionality values decay exponentially with strain to become zero at the steady-state condition.This behavior results from an underlying stochastic process in which load-carrying particles in the shear zone move into the steady-state at random shear strains, causing the number of load carrying particles in the shear zone which are not in the steady-state to decay exponentially with shear strain.This note proposes a reason for why load-carrying particles move into the steady-state configuration at random shear strains, and for why - as the number of particles in the shear zone that are not in the steady-state decay exponentially - so too does their total inter-particle contact area.The reason is that all load-carrying particles in the shear zone move at random during shear.If this is the case, then further entropic considerations show that the percent number of load-carrying particles must distribute exponentially against inter-particle contact area for load-carrying particles not in the steady-state and uniformly for those in the steady-state.

The stability of slip blocks on arch dam abutment under earthquake is of great importance for the safety of the dam. In order to analyze static and dynamic stability of arch dam blocks, a hybrid method incorporating the advantage of finite element method and rigid body limit equilibrium method is proposed. A scheme is presented by which any arbitrary slip block defined can be obtained from topography of the studied region, and then the 3D slip block is meshed. Based on stress field of static and dynamic FEM calculation, the stress field of all blocks is obtained by interpolating stress on block mesh. By integrating stress on block faces and using the idea of limit equilibrium method, the safety factor of slip block is computed and dynamic stability of arch dam block is evaluated. The proposed method can provide instant stability of slip block during the whole seismic action, which cannot be obtained with traditional limit equilibrium method. Taking Shapai arch dam as an example, the dynamic stability of Shapai arch dam abutment under Wenchuan "5.12" earthquake is analyzed. The method gives a new way for analytical technique of dam abutment stability evaluation.

In a Cam-clay model, the energy equation assumption is lack of theoretical basis and it cannot describe the dilatancy of sand. Moreover, it needs to determine the plastic potential functions. In order to improve the Cam-clay model, mathematical principle of the energy equation assumption and the dilatancy equation in the Cam-clay model were analyzed and a similar Cam-clay model for sand was proposed based on the generalized potential theory. In the proposed model, the mathematical principle of each assumption is clear and only four conventional soil parameters are needed. The difficulties brought by the determination of plastic potential functions are avoided. Rationality of the proposed model was verified by the test results of sand under different stress paths. The results show that the predicted results of the proposed model agree well with the test ones and the dilatancy of sand can also be described. It proves that the proposed model is more reasonable than Cam-clay model.

This paper aims to study the effect of long-term soaking and leaching on the stability properties of fine-grained soil with and without gypsum content, in relation to changes of the mechanical properties and permeability. The soil samples were stabilized with 3% lime and cured for 28 days at 20°C. The cured samples were subjected to soaking for different period extended to 180 days. Also, permeability values of stabilized and unstabilized soil samples were measured under 30 days of leaching. Results showed that the long-term soaking has strong effect on the unconfined compressive strength, wave velocity and volume change of the stabilized soil samples. These properties are degraded when compared with initial properties of the unsoaked samples. Moreover, the results indicate that an increase in gypsum content have significant effects on stability properties of soil samples. The leaching test revealed a marked increase in permeability values of unstabilized soil samples and, to a lesser extent, in the permeability of stabilized samples. The results also showed that the permeability of soil samples is highly related to the quantity of gypsum during the leaching test; however, the lower the gypsum content, the lower the permeability.

Rain is an important factor inducing the failure of slope. The existent studies concerning the influence of drainage way arrangement on the seepage flow in the slope are little involved. In this paper, the law of seepage flow in the slope exposed to the rain is investigated when the drainage way exists. The mathematical model for seepage flow of rain in the slope is set up, and the finite element method is used to solve the model. The numerical solution of the model is obtained. The results indicate that the drainage way arrangement within the slope has important influences on the distribution of pore water pressure in the slope. The arrangement with the drainage way in the slope and its end extending to the surface of the slope has better effect for the drainage; but, the arrangement with the drainage way within the slope along the strike direction of the slope is not a good one. The soil between the drainage ways is influenced weakly by the rain, which means that little rain seeps into part of the soil.

This paper presents an experimental investigation on the permeation process of grout inside soil. The injection of grout inside soil and its permeation process is not well known due to the fact that natural soils are not transparent.In this study, a transparent soil is used to visualize the permeation process. Transparent soil used in this study is made of fused silica and calcium bromide solution with the same refractive index. Geotechnical properties of transparent soil were found to be similar to those of natural sands. A combined grouting and optical measurement system is developed for this study, which consists of a grout injection station with a constant pressure head to inject grout into a transparent soil model, a laser to illuminate the interested cross-section inside the model, and a charge-coupled device (CCD) camera to capture a series of images during the whole grout injection and permeation process. Image processing techniques are used to delineate the fronts of grout body. It was found that the radius of grout body is proportional to the grouting time, which verifies Maag's permeation grouting formula.

High-water-content and quick-setting stones serve as one of the filling materials in the underground goaf, and their water-cement ratio and curing time have significant influence on the main mechanical characteristics. The internal structure of stones was analyzed by scanning electron microscope (SEM), and MTS815 Flex Test GT was used to determine the mechanical parameters. The influence of curing time and water-cement ratio of stones on their main mechanical characteristics was analyzed. The results showed: (1) the microscopic morphology of stones was structure of columnar sticks and mesh; (2) with an increase in water-cement ratio, the internal structure of stones became looser and the compressive strength became lower; (3) with curing time increasing, the strength of stones grew rapidly at the beginning, until reaching its maximum at the curing time of 7 days, but then decreased gradually and finally stabilized; (4) the variation characteristics of elastic modulus and deformation modulus were similar to that of compressive strength; and (5) although there was a small rise at the beginning, Poisson's ratio remained stable on the whole.

Electrokinetics has been applied in geotechnical engineering, specifically in soil improvement, for nearly a century.Most engineering applications and research projects have focused on electro-osmotic consolidation generated by the negative pore water pressure, as well as settling of suspended soil solids due to electrophoresis.Although many studies have reported that significant cementation bonds have developed in soils after electrokinetic treatment, a systematic study on the subject (i.e. the mechanism and observations from field and lab experiments) has not been reported.In this review, the cementation developed in marine sediments and offshore calcareous sands is presented along with discussion of the principles of electrokinetic cementation.The study is of particular importance in earthquake engineering for development of soil improvement technologies against seismic loading, because electrokinetics generated cementation of soils may significantly reduce the risk of soil liquefaction.

Rockburst is one kind of dynamic geological disaster, and the influencing factors are complicated and diversified. Among which the high geostress is the major external cause, the hard brittle rock mass is the major internal cause, and excavation disturbance is the major incentive. The initial horizontal geostress in the surrounding rock significantly affects distribution of secondary stress after excavation, which is closely related to rockburst occurrences. Therefore, it is necessary to study the excavation disturbance on rockburst trigger under different horizontal geostress. To accomplish the research goal, the underground powerhouse at Pubugou hydropower station in Sichuan Province, China was taken as an example. The numerical model was set up to simulate the actual excavation steps and predict rockburst intensity based on stress criterion. The analysis indicated that the potential hazard of rockburst increased with the rise of horizontal geostress. When horizontal geostress was more dominant, rockburst usually happened on the roof and bottom of the cavern. When vertical geostress was more dominant, rockburst usually happened on the side wall of the cavern. Specially, when the ratio of horizontal geostress to vertical geostress was greater than 1.7, we should seriously pay attention to the effect of excavation disturbance on rockburst trigger and take proper measures to prevent and control rockburst.

This paper examines the response of metro stations adjacent to an excavation via a series of comprehensive finite element parametric studies. The effects of several key factors (e.g., the distance, D, between the existing metro station and the deep excavation, the depth of the excavation, H_e) governing the performance of the metro station were quantified. On the basis of the numerical analysis results, some major findings were obtained:(1) rotation of rail closer to the excavation transformed from the clockwise direction to the anticlockwise direction with the increasing of D / H_e; (2) as D / H_e increased, internal structural forces of juncture at roof slab and right wall decreased gradually and excessive internal structural force occurred when D / H_e was smaller than 0.7; and (3) conventional protection criteria of metro stations were not rational because they did not take into account the internal structural forces.

It is well established that soils exhibit non-linear behavior even at small strain levels. Yet, most evaluations of the seismic response of soil deposits utilize an equivalent linear methodology, i.e. elastic solutions incorporating constant damping. In such solutions the modulus and damping constants are adjusted by means of an iterative approach to correlate to the maximum strain. In doing so, the smaller amplitude, high-frequency component motions are forced to use the same modulus degradation and damping as that of the lower frequency motions. As a consequence, the computed motions at the surface of a deposit often exhibit unrealistic low amplitudes at high frequencies when the strong input motions are applied. This article presents a modified Kelvin model, in which the modulus and damping are treated as loading history dependent coefficients for each loading-unloading branch. This model works in the program FLAC in a time domain integration procedure. Based on the peak strain level experienced in the previous half cycle, the modulus degradation and viscosity are updated for the current branch. Numerical analyses for a soft soil site are presented and compared with those obtained using the equivalent linear method implemented in the program SHAKE for a recorded motion at Treasure Island. The site response analysis of another soft clay site for a higher design input motion shows that the computed PGA and response spectral could be higher than those obtained from a typical equivalent linear analysis.

The initialization and development of strain localization inside a triaxial sand specimen are affected by multiple factors (e.g., membrane stiffness, weak inclusions, particle breakage, slenderness ratio, etc.) To investigate the effect of membrane stiffness, a flexible membrane algorithm was implemented in the discrete element method code (PFC), which allowed both vertical and radial deformations of the specimen. The modeling results showed that the specimen failed with bulging in the middle portion (i.e., presented a barrel-like shape). In comparison with the rigid membrane algorithm, prominent shear bands could be spotted at microscopic scale for the flexible membrane bounded specimen. Weak inclusions and particle breakage were simulated by reducing particle stiffness and changing particle shape, and their effects on the triaxial sand specimens were examined. Analyses on the modeling results indicated that - in addition to the primary shear bands - both weak inclusions and particle breakage intrigued a secondary shear band and shear block while imposing no substantial effect on the stress-strain response of the specimens.

The 2008 Wenchuan earthquake caused severe damage by landslide along the Longmenshan thrust belt zone in southwest China. The authors investigated various influential factors on slope stability and landslide mobility based on 97 landslides in Wenchuan County and 46 long runout landslides in Wenchuan earthquake area, respectively. The results demonstrated that slope height, horizontal peak ground acceleration (PGA), and geological structure were more influential to slope stability than slope angle and rock type. After slope failure, slope height, rock type, field cross angle and landslide volume were the most important influential factors on landslide mobility, and followed by slope angle and horizontal PGA. Two empirical models were developed to estimate landslide volume and landslide mobility. Their validity had been verified by the satisfactory agreements between observations and predictions. Therefore, they could be useful in the Wenchuan earthquake area and areas with similar geological conditions.

The results of dynamic triaxial test of reinforced geogrid cohesive soil under cyclic loading indicate that the dynamic strength of reinforced cohesive soil shows good regularity with the change of amplitude of dynamic stress and number of reinforced geogrid layer. The forms of dynamic strength curve of reinforced cohesive soil are basically uniform with different conditions, and the dynamic strength curve can be expressed as τ_d = A Ln (-B Ln N_f). The axial cumulative strain increases quickly with vibration times increasing under cyclic loading during the earlier stage of the vibration which can reach 60% ~ 70% of the total strain. It remains stable with vibration times increasing if the dynamic stress is smaller than the limited dynamic stress. However, the axial cumulative strain increases slower than the axial cumulative strain during the earlier stage, and the curve of the axial cumulative strain is totally nonlinear if the dynamic stress is bigger than the limited dynamic stress. The sample eventually fails for instability. The development of the axial cumulative strain of reinforced cohesive soil shows good regularity along with the change of amplitude of dynamic stress and number of geogrid layers. The cumulative strain may be described as ε_p = A + B N ^C.

The problem of seismic liquefaction related to artificial fill embankment is complicated. In this study, by making using of Finn model, a typical artificial fill embankment was simulated by means of Finn model in FLAC. Numerical simulation results indicated that the pore water pressure of saturated sand soil increased extremely which resulted in the significant decrease of the effective stress. When the effective stress decreased approximately to zero, the liquefaction phenomena occurred. According to the change of pore water pressure at different locations and depths, three rules have been obtained. Firstly, the occurrence time for the first peak of pore water pressure coincided with the seismic peak acceleration of input wave. Secondly, the liquefaction occurred earlier in the upper layer of saturated sand soil than in the lower layer below the bottom of slope. However the duration of liquefaction in the lower layer of saturated sand soil was longer than in the upper layer. Thirdly, the upper layer of saturated sand soil was almost not liquefiable below the top of slope, while the lower layer was easily liquefiable with long duration time. In summary, the characteristics of seismic liquefaction in the artificial fill embankment were closely related to the seismic wave, locations and depths of saturated sand soil in a slope.

Hardin-Drnevich, Stokoe and Davidenkov models all belong to hyperbolic models which are commonly used to describe the relationship between dynamic shear modulus ratio versus shear strain of soils. However, few studies have been performed on the parameters of these models. To understand the meaning of the parameters, the three models were analyzed by using test data from remolded, disturbed and undisturbed soils. Then, by using fitting approach, the parameters of the three models were derived. After comparing the fitting parameters with the reference strain γ_r from dynamic triaxial test, it is found that the fitted parameter γ₀ in the Davidenkov model has same meaning with the reference strain γ_r. Furthermore, the factors of impacting the reference strain were performed by changing confining pressure and loading rates and the effects of the reference strain on dynamic shear modulus ratio were discussed. By using the results derived from the study, a range of reference strains for clay, silty clay, silt, muddy clay and sand was recommended.

Waste tire mat-like geo-cells can be used as containers filled with soil and planted with vetiver grass. Waste tire and vetiver grass root reinforced soil are unxial and triaxial compressed to study the strength and deformation behavior. Test results indicate that tires and vetiver grass root can increase shear strength and improve ductility of soils. An in situ demonstration test is also conducted by waste tire mat filled with soil and vetiver grass in bioengineering stabilization of a fully weathered mud rock slope. Nonlinear viscosity of the slope soil is taken account into Bingham model that indicates the composite slope soil is fully or partially saturated. The concept of shear layer can be easily emerged by the formula without any assumption. As a result, the progressive deformation of the waste tire mat and vetiver grass root reinforced slope soil is described.

The compression modulus is a key parameter for evaluating compression behavior of soil layer and for predicting foundation settlement. High-pressure confined compression testand the standard penetration test were performed to investigate the compression characteristics of sandy soil layers beneathpile toe, and to obtain the relationship of compression modulus with depth and overburden pressure. The experimental results indicated that the compression modulus of sandy soil increased with pressure, but its increasing rate gradually reduced. The compression modulus almost linearly increased with embedded depth.The ratio of compression modulus under the pressure ranging from 100kPa to 200kPa tothat under the pressure in the range of overburden pressure to the additional stress plus overburden pressure (Es/Es_0.1-0.2) ranged from 2.1 to 4.2. The empirical formula of the Es/Es_{0.1 -0.2} ratio with overburden pressure was established to give a reference to predict compression modulus of sandy soil under actual pressure, as its overburden pressure was known. At last, the basic principles of how to determinate the value of compression modulus were proposed for design of pile foundation.

To investigate the effects of shear ring, drilling mud skin and anti-corrosion coating on force and deformation characteristics of Steel Tubular Composite (STC) piles, seven model tests with different combinations of shear ring, drilling mud skin and anticorrosion coating were conducted. Based on test results, the relationship between the shear ring spacing and STC pile bending stiffness is studied. The results obtained from the tests show that the capacity for lateral deformation of STC piles is greatly influenced by drilling mud skin, anti-corrosion coating and shear ring. Drilling mud skin and anti-corrosion coating reduces the bond strength between steel tube and concrete. On the contrary, shear ring can effectively transfer the stress from concrete to steel and, consequently, the bearing capacity for lateral deformation of STC pile is improved. This also causes the strain distribution along the cross-section to be "inverse-v" or "m" shaped. The reinforcement of shear ring can completely overcome the adverse influence of drilling mud skin and anti-corrosion coating, when the spaced distance is less than 2D (D is pile diameter). Finally, the relationship between the increment of bending rigidity of STC pile and shear ring spacing is formulated.

This paper has concentrated on technical effectiveness of Chemical-Physical Combined Method (CPCM) through a series of laboratory tests. The chemical stabilization method was used to investigate two slurry groups, one initially with lower water content w (w<2LL; LL-liquid limit) and the other with higher initial water content (w≥2~6LL) which is within the water content range of dredged clay slurry. The final water content, shear strength Cu of clay slurry achieved and their relationship after chemical pre-treatment were set up for the two groups at different chemical dosages, while some vacuum preloading (VP) tests using pure clay slurry were also conducted for comparison. When the initial water content is lower (w<2LL), the achieved shear strength with the chemical treatment can be higher than that of pure clay slurry treated by VP with drains. However the shear strength after the chemical treatment is generally poorer when the initial water content is higher (w≥2~6LL). In order to achieve better or higher performances, the VP with drains was introduced during the chemical stabilization. The achieved higher strengths with this combined method have proven that CPCM for improvement of clay slurry is not only academically effective but also very potential for engineering applications.

For high earth core rockfill dam constructed on thick overburden, traditional anti-seepage system design and construction is to construct the foundation seepage cut-off wall first, and then build a rigid or soft-connected grouting gallery on top of seepage cut-off wall, and then cover the gallery with high plastic clay. Therefore, after the construction, the high dam body transmits great pressure to the seepage cut-off wall through the gallery. For this reason, a new type of structure for high earth core rockfill dam constructed on thick overburden is proposed, namely constructing a "concrete slab" with a large dimension gallery on the overburden so that the seepage cut-off wall can be constructed from the inside of the large dimension gallery during or after construction of the dam body to reduce the stresses in the cut-off wall. The cut-off wall and the concrete slab on the overburden, together with the upper dam body, form a complete anti-seepage system; but, it has not been used in actual dam construction. In this paper, two-dimensional finite element stress and deformation analyses were carried out to assess the effects of the proposed concrete slab and cut-off wall construction sequence on a 223-meter-high earth core rockfill dam constructed on thick overburden. The stresses and deformations of the concrete slab are obtained, and its structural safety and suitability for use in earth core rockfill dam are also evaluated. According to research results, by rationally adjusting the construction order of the dam body and the cut-off wall, the compressive stresses on the cut-off wall can be controlled, which has been a big problem for traditional structures. It reflects the advantage of the new structure.

Fissures generally exist in all of expansive soils. It always results in failure of expansive soil slope. Shear strength characteristics of fissure plane are studied and a new triaxial test method is recommended with typical expansive soil taken from Nanyang section of middle route of the South-to-North Water Transfer Project (SNWTP). The computed tomography technology is firstly introduced to shear strength test of fissure plane. The real failure stress of fissure plane is analyzed accurately through measurement of the fissure plane's real occurrence by CT scanning and shear strength parameters determined method is provided for triaxial test of fissure plane. The test results show that peak strength of fissure plane is much lower than that of bilateral expansive soil without fissures, and even smaller than residual strength of bilateral expansive soil without fissures. The reasons of these phenomena are that gray clay in fissures has higher water content, smectite content and particles orientation degree.

It is very critical to understand the dynamic interaction of vehicle-track-subgrade for high-speed railway development. This paper presents a dynamic interaction model of vehicle-slab track-subgrade system, in which a slab track-subgrade structure is put forward as a substructure and is studied considering track irregularity excitation and displacement compatibility condition. In this model, both the actual structure of a vehicle and vibration characteristic of the substructure (including slab track, subgrade and foundation are) considered. This model not only reflects the interaction between vehicle and substructure completely, but also provides a theoretical foundation for determining design parameters of high-speed railway (especially for design parameters of track and subgrade). Research results can provide technical reference for design and construction of slab ballastless track and subgrade structure of high-speed railway.

Engineering characteristics of soil are different at dissimilar layers and regions because of the variability of geological origin, earth history, environment condition, etc. Traditional safety factor method, which is usually used in design and calculation of geotechnical engineering, cannot reflect this variability and uncertainty of soil scientifically because characteristics of soil are often influenced by sampling and experimental condition. In this paper, reliability analysis method, based on probability and statistics, was applied to analysis anti-overturning stability of supporting structure of excavation. Anti-overturning limit status function, in which layered soil was considered, is derived using equivalent beam method. In the process of computation, geometric dimension of supporting structure was deemed as constant such as excavation depth, anchored depth of pile, pile spacing and so on. Soil parameters, gravity, cohesive force, and internal friction angle of soil were regarded as random variables. At last, the open excavated tunnel of Chengdu-Dujiangyan railway was taken for example to discuss anti-overturning stability of supporting structure influenced by loads nearby excavation.

The vital geological risk factors including in saturated soft clay stratum, water-rich fracture zone, Karst stratum with abundant soil caves and saturated sand stratum on the safety of metro station are discussed. The successful experiences of controlling the geological risks in metro projects are summarized. The design methods of metro station foundation and main structure at geological sensitive areas are presented. It is concluded that hazards of foundation settlement, pit collapsing, and water gushing are more easily taking place in metro stations with saturated soft clay stratum. Failure of soil and water inrush often happened owing to the water-rich fracture zone. Collapse and subsidence frequently come up at the area of Karst stratum with abundant soil caves. The water bursting and sand flowing is usually broken out at the foundation pit with saturated sand stratum. The safety of metro station with the above-mentioned defective geological factors can be controlled and ensured through targeted design schemes and construction methods.

The water level in the reservoir of the Three Gorge Dam rises from 145m to 175m since the dam is in full operation. The periodical fluctuation of the water level changes the water environment of the reservoir banks. The water saturation and drying cycles affects the strength and stiffness of original rock and soil mass. These changes have led to problems such as the instability of ancient landslide and the safety of buildings supported on pile foundations whose bearing stratums are affected by the changing water levels. Seven 8-story height buildings supported on pile foundations in Fengjie county in Chongqing Municipality had been facing such threats. A series of field direct shear tests and pile loading tests were conducted to study the influence of capacity of foundation soils due to water level fluctuations. The results indicated that after water soaking, the cohesion strength of the foundation soil decreases significantly, while the internal friction angle is relatively stable. The weaken ratio of the internal friction angle which is the ratio of strength parameters before and after water soaking into the gravel soil is mainly between 0.78-1.0 and the weaken ratio of cohesion strength is around 0.4 to 0.8. The capacity of pile decreased when the water submerged the pile tip. Field calibration and indoor calibration have proved the availability of the O-cell boxes designed and produced by ourselves in pile static loading tests. The observations provided important support for evaluating the safety of seven 8-story height buildings.

The crag body at the slope of the Lvchunba tunnel entrance is so large that it's necessary that the dangerous area and minor dangerous area are determined first in the design. Then, the proper and economical treatment is adopted. There is no division method of dangerous areas in the current codes and standards. So, field testing is adopted for inverse analysis of the numerical parameters. After that, the rock-fall motion path is simulated for all of the sections by particle flow code. Last, the dangerous and less dangerous areas are determined according to the influencing range and degree of rock-fall on all of the sections. Based on the particle flow code, the 2-D visual simulation of rock-fall motion path is realized, and the motion style of free fall, bouncing and rolling can be obtained. Then. it's feasible to analyze the dangerous area of crag body. Also, the motion velocity and displacement of the rock-fall at any time can be gained so that the reference is presented for calculating the energy or impact force of rock-fall at different positions. The simulation method of particle flow code can be used as a reference for the design of similar railway engineering.

This paper presents a case history detailing the two concentrating belts of unloading fractures found in the right bank slope of Dagangshan Hydropower Station, China. Based on the analyses about potential sliding modes which controlled by the unloading fracture belts, the cause of the three in-situ crack cases are discussed. The results of numerical calculation illustrate the stability of excavation slope and rock mass failure trend, verify development patterns of the in-situ cracks and instability mode of hazard sliding blocks, offer references for reinforcement. The most appropriate and feasible method of stabilizing slopes with deep-seated fractures is to construct synthetic reinforced galleries system and combine them with grouting and prestressed anchoring.

Based on direct shear tests of soil samples with roots, the empirical formula of roots' contribution to soil's shear strength is summarized. In consideration of the effects of sliding surface depth, sliding mass width, sliding mass shape and roots, shallow slope stability analysis method for vegetated slope is suggested by mechanical equilibrium analysis of shallow sliding mass. According to the analysis of factors affecting shallow slope stabilization, it is shown that the factors' importance, in descending order, is sliding mass width, sliding surface depth and sliding mass shape, in turn. Results of shallow slope stability analysis proposed by the paper are reasonable in comparison to the plane analysis method because more factors are taken into account.

In order to study the settlement characteristic of ram-compacted piles with bearing base (RPBB), pile-raft composite foundation in high-speed railway, RPBB, and Cement Fly-ash Gravel (CFG) piles were tested and compared under the Beijing-Shanghai high speed railway. Using hydrostatic levels, the settlements of test sections were monitored over the long term. Results show, because of the formation conditions and the construction technology of RPBB, the reinforced depth of RPBB is limited. Consequently, the settlement of RPBB is bigger than CFG. It also takes longer to reach stable settlement for RPBB. In addition, based on measured and modified soil compression modulus, the thicknesses of compressed layers and the settlements were calculated using control strain method and control stress method. The calculated results using control strain method and modified soil compression modulus are reasonably close to the testing data. This proves that the modified soil compression modulus can improve the precision and reliability of the settlement estimation. This can overcome the limitations of existing Chinese code.

The long period spectrum characteristics of microtremor have been used in investigating dynamic parameters and speculating the graben structure. This paper reports the results from a microtremor observation profile crossing the Xuzhou City to deduce the graben structure of the Xuzhou Basin, and investigate the predominated period in the H/V spectral ratio method. When the equivalent shear wave velocity is constant, the result shows good corresponding relationship between predominant period and sedimentary thickness. The theoretical calculation of the microtremor characteristics based on wave propagation theory validates the analysis results from the observation. The method, imaging the graben structure based on microtremor with H/V spectral ratio is credible, and the predominant period obtained from H/V spectral ratio is more precise than horizontal power spectrum.

The fundamental construction of disk type p/s composite piezoelectric transducer, which is capable of measuring both shear and compression wave velocities simultaneously in triaxial testing conditions, is discussed. Performance evaluation of wave measurement system was conducted by studying the effects of crosstalk, quality of transmission cables and features of recording unit. The evaluation showed that crosstalk deterioration can successfully be eliminated through proper grounding, and the use of low-noise coaxial cables can significantly enhance signal-to-noise ratio of received signals. Use of wave recorder with low inherent noise, and high analogue to digital (A/D) conversion resolution, sampling rate, and voltage sensitivity can considerably improve the overall performance and signal interpretation of received signals. Combined monitoring of shear and compression waves by using disk transducer was also found to assist the identification of near-field effects in shear wave receiver signals. Based on this work, disk type piezoelectric transducers show significant future potential for laboratory determination of shear modulus and constrained compression modulus of soil because of their robustness and noninvasive nature.

The differences of boundary restraints directly affect the dynamic responses and bearing capacity. The mechanics models of beams with flexible supports were established, and elastic supports, lumped masses at the beams ends and non-symmetric supports were taken into account. Numerical analysis of the model was conducted by using the Euler-Bernoulli's theory and Lagrange method. The characteristic equation and the eigen-functions of the supported beam were derived. The responses of beams subjected to any dynamic loads were explored. Through the comparison with results from the existing references calculation results, the accuracy of the present method is validated. Then, the effect of the boundary restrained stiffness on the dynamic responses of beams is analyzed. The analyses show that flexible supports have great effect on the performances of beams, both horizontal and vertical elastic supports can reduce the dynamic coefficient of displacement. The optimization of added masses and their positions on the beam can improve the impact resistance of beam.

Shafts play an important role in subway construction. Some shafts will be retained and used as an important part of the permanent subway structure after the excavation. More knowledge about the deformation mechanism of shaft construction is needed as it has a great influence on the safety of the subway construction. Based on a shaft construction of Changchun Metro1, this paper systematically analyzes the impact of various factors on the enclosure structure and its surrounding areas by examining a large number of monitoring data obtained in real-time during the shaft construction. The results show that the inverted side-wall method is appropriate for shaft construction under complex conditions, as this method of construction would provide a good control over the deformations of the shaft enclosure structure and the surrounding layers. The real time monitoring data show that the ground settlements during shaft construction are greatly influenced by local loads and construction traffic applied on the ground surface; e.g., the ground adjacent to the shaft loaded with excavation muck piles settled significantly more than the other side that was less loaded or not loaded. In addition, rapid change of groundwater level during construction dewatering could also significantly impact the shaft structure and its surrounding environment, especially for the surrounding ground settlement. The result of this study could provide some useful information for the design and construction of similar engineering structures.

This paper took the excavation of gravity type anchor ingot deep foundation of Nan-xi Yangtze River Bridge as the research object. A typical section of the northern slope was selected to make a 3-D numerical model. Through the equilibrium method and the numerical calculation software FLAC^3D, dynamic stability of the slope, before and after being reinforced by bolts, was calculated and analyzed. Both the stress and displacement variations of the slope were compared, and the dynamic response characteristics of the slope and bolts were explored. The results show that both the maximum tensile stress and shear stress of the slope decreased, the dynamic disturbance degree of the slope inner layer mitigated observably, the latent sliding plane migrated toward the empty face, and the dynamic stability safety factor increased when the slope is supported by bolts. However, as the head of the lowest bolt located in the fine sand layer and close to the explosive source, it will shake more intensively and cause partial mortar damage, and the reinforcement effect of bolts decreased.

Sand and cobble stratum in Chengdu Metro Line 1 presented challenging tunneling ground conditions with high water levels, and high strength large diameter cobbles that resulted in excessive wear of tunnel boring machine cutters. In general, the stratum is self-supporting for extended periods of time, making real-time monitoring for ground subsidence difficult to detect. In the process of tunneling, the traffic along roadways over the tunnel bore had to be restricted because of ground subsidence or multiple ground collapses, which have serious implications on the surrounding environment. From some ground collapse cases, the excavation parameters and real-time monitoring data were investigated to understand the ground subsidence mechanism. The research shows that: 1 Under the condition where the stratum has high water levels, non-uniform densities and higher permeability coefficients, the sand and cobble soil is in a plastic state. 2 Alone, the stratum cannot maintain earth pressure balance for EPB tunneling. 3 Additionally, once the stratum is disturbed by dynamic loads, the self-supporting nature of the ground degrades progressively. 4 Finally, the subsidence manifests at the surface as ground collapse. Our study further investigates the major factors of ground subsidence induced by EPB tunneling in such stratum. In order to prevent ground subsidence, some valid measures that were verified by previous projects were advised. The results are helpful for improving the understanding of the mechanism for ground subsidence, avoiding ground collapse and environmental impacts, and maintaining surface safety.

Underground excavation will influence the safety of the geotechnical surroundings. Based on the stochastic medium theory, influence on the surrounding rock and soil by underground excavation is studied. Formula for predicting movement and deformation of rock and soil mass caused by excavation are deduced, and especially for those with horizontal rectangle section, semi-analytic solution for this problem is put forward. Furthermore, a computer program is developed to compute the movement and deformation of the rock and soil mass resulting from excavation. The theory is applied in an engineering example, which shows that the results of the research are reasonable and the developed solutions can be used for safety evaluation of underground excavations.

Traditional Terzaghi foundation bearing capacity superposition empirical formula does not consider the nonlinear effect of the soil unit weight on the ultimate bearing capacity and its failure mode, which is only a kind of approximate method. Therefore, error occurs when unit weight is taken into account. The relatively complete, two-dimensional conversion velocity field based on slip-line field and the basic method and principle of corresponding failure mechanism were introduced in this paper. Compared with the existing upper bound failure mechanism, this failure mechanism adds a one-dimensional optimization freedom degree, which can reflect the real failure mode, and thus obtain better upper bound solutions. The results indicate that the traditional superposition empirical formula stress on safety, and the error caused by traditional superposition method will increase with the increase of the internal friction angle. The failure range derived from non-superposition method is between the failure range determined by using superposition method to solve the analysis coefficient N_γ and N_c (or N_q), which reflects the influence of soil unit weight on the failure mechanism.

Landslide deposits are geologic features that show signs of movement ("failure") which occurred outside the historical record. Unlike recent landslides, they do not show evidence of recent movement such as a defined head scarp, tension cracking, and/or bulging at the toe, but are identified by geomorphic evidence of past movement. Back analysis of landslides is a common technique used to assess shear strengths for design wherein a factor of safety (FS) equal to 1.0 is assumed along a pre-defined failure surface. However, as landslide deposits may no longer be "at the verge" of failure, a back analysis with an assumed FS equal to 1.0 may result in excessively conservative shear strength properties. Therefore, back analysis of landslide deposits should be performed assuming a FS greater than 1.0. As there is no consensus on how much higher than unity FS should be, values typically used in practice range between FS = 1.05 and 1.25 with little, if any, justification provided for FS selected. Consensus on seismic FS and/or calculated maximum permanent seismic displacement for use in back analysis is similarly lacking. In this paper we evaluate parameters that affect back analysis of landslide deposits and provide illustration using a recent case history from our practice.

Dynamic penetration test (DPT) is one of those widely applied to in situ test techniques. The rod length during operation could be as large as hundreds of meters, calling for length modification in practical applications. Two major methods were proposed in the most recent specifications for the modification of dynamic penetration rod length. These two methods were based on Newton elastic collision theory and elastic rod waving theory, respectively, and derived totally different correction coefficients. How to apply rod length modifications to the results of dynamic penetration tests has become a major issue that affects the reliability of its application. In this study, indoor model tests were conducted to evaluate the length corrections. A variety of heavy dynamic penetration models was conducted on sand and gravel soil with different rod lengths (2.0 m, 8.9 m, 16.4 m, 23.4 m, 30.0 m, and 36.0 m). The modification coefficient of heavy dynamic penetration rod length was obtained from these experiments. The results indicated that the modification coefficient for the length of penetration rod followed the description of Newton elastic collision theory, and was independent of the material properties.

Based on shaking table test and finite element method, the influence of water on the natural vibration characteristics of the pier was studied. Then, the effect of long-period ground motions to hydrodynamic pressure of the pier under various water depths was discussed. As the same time, seismic responses of the pier were also analyzed under the condition of different water depths. Finally, the first order natural frequency and the distribution of acceleration peak values along the pier height were compared between the shaking table test and finite element method. The results show that the influence of the water on the natural vibration characteristic of the pier should be considered. When the water depth exceeds a critical value, the hydrodynamic pressure has a different tendency along the pier under different ground motions. The peak values of bending moment and shear force at the bottom of the pier and the acceleration and displacement on the top of the pier increase with the rise of water depth. The calculated values from model test and finite element calculation are in good agreement.

A fuzzy comprehensive evaluation method was put forward to study the seismic risk of tunnel portals. Through collecting and analyzing information of seismic damage of the tunnel portals after Wenchuan Earthquake, the main influence factors of tunnel damage and its function law were summarized. The risk factors' weight and the membership degree to the risk level were calculated respectively by AHP and trapezoidal membership function and semantic subordinate function. Seismic risk level of tunnel portals was concluded through fuzzy transformation principle and weighted average method. At last, a case was done on a built tunnel by fuzzy comprehensive evaluation. The conclusion could be used to take corresponding measures to reduce seismic risk.

In order to study the influence rule of the security and stability under different goaf dip angles during the tunnel excavation when the highway tunnel is approaching to the goaf, the computation models are built and simulated numerically with the finite difference software of FLAC, relying on the Tian Pingzhai tunnel of Dazhou-Wanzhou highway. Under the condition of the certain tunnel buried depth, thickness of coal seam and distance, a comparative analysis on the displacement and axial force of primary support resulted from the changement of the overlying goaf dip angle has been conducted. The impact on the displacement and deformation of surrounding rock for different goaf dip angles are also further explored. The simulation results indicate: (1) with the increase of the dip angle, the added forces acting on primary structure will decrease; and (2) when the dip angle is less than 5°, the impact on the added forces IS most significant. The changement of the overlying goaf dip angle brings remarkable impact to vault settlement and horizontal convergence, and the influence on left hance (which is close to the mined-out area) is greater than the right hance, but it has little impact on the displacement of arch bottom. It is suggested that Pre-firming and Treatment for mined-out areas should be used.

Failure of a landslide dam might occur by river discharge or rainfall. A rise of the upstream water level of a landslide dam and rain infiltration into the dam body increase pore water pressure and the weight of the dam. In this study, transient seepage analysis of rainfall infiltration and dam stability analysis are performed. A two-phase flow simulation using the FLAC finite difference code is adopted to analyze unsaturated seepage flow in transient fluid-mechanical calculations. The safety factor of dam stability is evaluated using the shear strength reduction technique. The parameters discussed in this study include the rising speed of the water level, rain infiltration, and the hydraulic conductivity of soil. The results show that the time to slope failure of the dam is approximately 247 min. when only the effect of the rising upstream water level of the dam is considered. The failure time decreases to 189 min. when the rain infiltration and rising upstream water level of the dam are considered. The results also indicate that the hydraulic conductivity of the dam affects dam failure time.

The effects of moisture content on strength and deformation characteristics of gassy muddy clay with thin silt interlayers obtained from a tideland reclamation area in Cangnan are analyzed by conducting triaxial and consolidation tests. Results show that the moisture content of the clay significantly affects the shear strength of the soil. The total cohesion of the clay presents a ladder-type steep reduction as soil moisture content increases, while the internal friction angle is almost free from the influence of moisture content. Soils with different moisture content exhibit relatively linear elastic properties at a small strain. The stress-strain curve alternates between mild strain hardening and mild strain softening as soil moisture content varies. The overall shear strength of the soil is determined by the combined action of the muddy clay and the thin silt interlayers, but the strength of the interlayers is more sensitive to changes with variation of the average soil moisture content than the muddy clay. The overall deformation characteristics of the soil are mainly determined by mechanical properties of the muddy clay. The thin silt interlayers and the average moisture content slightly influence the soil deformation.

Material properties of horizontal sandstone and mudstone interbedding rock mass have significant differences in the horizontal and vertical directions. It is difficult to reveal the characteristic of surrounding rock displacement induced by tunnel excavation in horizontal interbedding rock mass by numerical methods based on isotropic constitutive models. Regarded as transversely isotropic medium, material parameters such as density, elastic modulus, poisson ratio etc. of interbedding rock mass are deduced. As an example, transversely isotropic constitutive model is taken into consideration in three-dimensional numerical model of Luan Jiayan tunnel on Guangnan highway. The process of tunnel excavation is simulated in this model, and ground displacement distribution and variation with excavation steps are obtained. For the purpose of comparison, field tests of vault settlement and side wall convergence caused by tunnel construction are carried out. It is shown that, compared with isotropic constitutive model, transversely isotropic constitutive model can reflect the physical difference in horizontal and vertical directions for interbedding rock mass of sandstone and mudstone. Transversely, isotropic model is elastic and plastic zone caused by tunnel excavation can't be obtained. So, there exist some differences from actual situation when this model is applied.

A transversely isotropic Cosserat continuum model is developed based on the Cosserat continuum theory and transversely isotropic elastic theory in three-dimensional condition. The finite element program based on this model is developed to analyze the stress and deformation of transversely isotropic layered soils subjected to a loading on an elastic rectangular plate. It illustrates that the transversely isotropic model can reflect the soil's property more realistically than the isotropic one in some practical engineering, and that the proposed transversely isotropic Cosserat continuum model can describe micro-scale mechanical property of soils and simulate the mechanical behavior of the transversely isotropic soils effectively.

A plasticity model for simulating 3-D hysteretic behavior of frictional materials is presented. The model, implemented in the program FLAC^3D, is a generalization of the 1-D parallel Iwan model, in which a number of components are connected in parallel, each component consisting of a linear elastic spring in series with a perfectly plastic slider. This simple arrangement produces non-linear stress-strain curves and hysteretic, Masing behavior for 1-D cyclic loading.Remarkably, this approach also works for 3-D cyclic loading. For generalization to 3-D, the constitutive model consists of a number of 3-D linear-elastic perfectly-plastic components with Drucker-Prager yield conditions. All components are subjected to the same strain, with the global stress defined as the average of the component stresses. This simple framework is straightforward to implement numerically, leading to a robust tool for analysis of general 3-D stress paths. The 3-D Masing behavior of the Iwan model is demonstrated for simultaneous cyclic loading in two and three orthogonal directions. The interaction between stress-strain loops in orthogonal directions is explored. Further work is required to validate the predicted interaction against laboratory experiments.

The advances on soil-pile-structure interaction are described in this study. The nonlinearity of soil is accounted for approximately by a boundary zone model, and the curves of stiffness and damping of the nonlinear soil-pile system are provided. The coupled horizontal and rocking vibration of an embedded foundation (including pile cap) is analyzed by four parameters rather than the traditional six parameters. The radiation damping is corrected based on many dynamic tests in the field. The effects of soil-pile-structure interaction on dynamic behaviour are examined based on an engineering case. Three conditions are considered: (1) the soil-pile-structure interaction is accounted for fully; (2) the soil-pile system is flexible, but the structure is assumed to be rigid; and (3) the structure is flexible, but the base foundation is assumed to be rigid. For practical applications, a tower structure supported on piled foundation was examined under seismic loads. The earthquake forces and response were calculated using the time history analysis and response spectrum analysis, and compared with those using the method of equivalent static loads.

One of the most important features of flexible retaining walls is that wall deformations highly influence the distribution of earth pressures on the wall. The deflection of flexible retaining walls is controlled by the flexural rigidity of the wall and the soil pressure generated based on the permissible movement of the wall. The conventional method of design for flexible retaining walls is based on a limit-equilibrium theory that searches the force equilibrium by assuming the soils engaged around the wall are all in the limit state. The limit-equilibrium analysis does not take into account the nonlinear mobilization of soil reaction with wall deflection. The application of soil-structure interaction (SSI) to the design of flexible retaining walls means that the deformation of the structural system is analyzed based on the mobilized earth pressure and soil resistance along the wall. Such analyses will result in the most appropriate selection of the size and configuration of the wall, and for the tieback, while ensuring that deformations throughout the system are acceptable. The method introduced in this paper is to model the structural elements in terms of overall behavior and to use nonlinear p-y curves for modeling the passive resistance of soils due to lateral deformation of embedded wall sections. This paper will discuss the modifications of p-y curves that are needed to take into account the configuration of the wall system (group effects), the unsymmetrical driving forces in the backfill side, the soil resistance in the penetration side, and the long-term effect from the sustained loads.

This study analyzed resilient modulus (M_r) and other routine properties [gradation, LA Abrasion loss, standard Proctor, and unconfined compressive strength (UCS)] of commonly used limestone and sandstone aggregates in Oklahoma. A total of four stress-based regression models were evaluated using a statistical software package ("SPSS", Version 17), and material constants (k₁, k₂, and k₃) for these aggregates were determined. The octahedral model was found to outperform the other models and is recommended for use in Level 1 analysis. Correlation equations for material constants, required for Level 2, were developed using routine aggregate properties, and the universal model was found to be the "best fit" model with a R² value of 0.57. The estimated M_r values obtained from different models were in agreement with each other, and the variations of M_r values were within 4%. However, all of these models would result in conservative designs compared to the MEPDG recommended typical values. The findings of this study are expected to be helpful in the implementation of the MEPDG in Oklahoma and elsewhere.

A statistical analysis of the displacement monitoring data, the rock bolt stress and the joint meter opening degree during construction has been performed based on the monitoring data of five large-span rock-anchored beams of underground powerhouses, such as Pubugou Hydropower Station. It is proposed to classify the displacements of rock-anchored beams into large, medium and small types based on the average displacements of greater than 20 mm, 10-20 mm, and less than 10 mm, respectively. When the average displacement of the rock-anchor beam is within the large displacement range, pre-warning shall be given. The rock bolt-joint meter correlation pre-warning method is put forward here, which says that the maximum allowable deformation of the rock bolt is derived from the design tensile strength of the rock bolt and the unanchored length of the rock bolt and compared with the joint meter opening degree monitoring data in actual projects. The results show that the allowable deformation range coincides with the rock bolt deformations in different projects. The calculated allowable deformation can be regarded as a very important reference value in pre-warning of the joint meter opening degree. The rock bolt-joint meter correlation pre-warning method may be used in similar projects.

The ultra-soft slurry exhibits significant, initially viscous and subsequently elasto-plastic consolidation behavior. In order to investigate the different consolidation response at the viscous stage (VS) and elasto-plastic stage (EPS), a series of element consolidation tests were carried out in the modified light-capped oedometer. Test results show that the consolidation coefficient and permeability at VS and EPS vary significantly, which would bring in non-negligible error to the consolidation prediction with the ordinary assumptions. A simplified 1-D consolidation model considering the significant behavior difference in the VS and EPS was developed. The suction pressure proposed by Hong (2007, 2010) was employed to identify the consecutive stages in the model. Then, the consolidation tests for ultra-soft clay under different loadings were conducted in the small-scale consolidometer and the test results are simulated with the proposed model. The model can describe the progressive consolidation well from the VS to EPS. It provides an easily-accessible and reasonable alternative for the prediction of the consolidation behavior for the ultra-soft slurry.

Offshore piled raft foundations are one of the most commonly used foundations in offshore structures. When a raft foundation alone does not satisfy the design requirements, the addition of piles may improve both the ultimate load capacity and the settlement performance of the raft. In this paper, the behavior of a piled raft on undrained soil is studied based on a series of parametric studies on the average and differential settlement of piled raft using three-dimensional finite element analysis. The settlement behavior is found to be dependent on the number of piles and raft thickness.

Padma River is 6 km wide at the location of the proposed bridge.The estimated riverbed scour is 62 m deep with a return period of 100 years. Due to the proximity of Bangladesh to the tectonic plate contact zones of the Himalayas, the site is prone to seismic events. The deep scour and seismicity made the foundation design very challenging. A detailed, state-of-the-art ground investigation was carried out down to a depth of about 160 m in the predominant micaceous sands encountered to establish the static and dynamic soil parameters. Undisturbed samples of sand were recovered using the GEL PUSH sampling technique. State-of-the-art insitu tests, the self-boring insitu friction tests (SBIFTs), cross-hole seismic shear wave and compression wavetomography, self-boring pressuremeter tests, and flat plate dilatometer tests were carried out. Cyclic triaxial loading tests, bender element tests, k₀-consolidated stress path compression and extension tests were carried out. A structural numerical model developed using MIDAS software comprising six 150-m-long spans was developed including the piles at each pier. Using the dynamic design soil parameters, the ground response was determined using SHAKE, with strong ground motions corresponding to a 475-year return period earthquake event. They were then input into the MIDAS dynamic soil-structure interaction model with the Penzien Method to design the piles and the superstructure, taking due note of the deep design scour depths. The analyses demonstrate that six 3-m diameter steel raking piles driven down to about 120 m are needed to support each bridge pier with the 150-m span truss girders placed on seismic isolation bearings.

Many tunnels were severely damaged during high-intensity earthquakes in recent years, especially for those in mountainous areas. This paper analyzes the seismic responses of the circular deep grouting technique for mountain tunnels with surrounding rocks. Eight circular, deep grouting, pre-reinforcement schemes are simulated using 3D Finite Element Method. The deformation and stress characteristic and dynamic responses of the tunnel structure are analyzed. Results demonstrate that different grouting schemes should be applied for tunnel structure in the high-intensity areas. A new seismic performance evaluation approach is proposed in the paper based on the maximum relative displacements of adjacent locations on the tunnel lining. This evaluation method provides a better interpretation of the tensile and compressive states of tunnel structures that are subject to strong ground motion during an earthquake. The approach described in the paper provides guidance to seismic design and construction of tunnels in seismic, active, mountainous areas.

The Santa Clara Valley Water District (SCVWD) owns and operates the Calero Dam facilities, which include the Main and Auxiliary Dams. The Main Dam is about 98 feet high and straddles a former wide stream valley. A large part of the downstream embankment is underlain by coarse-grained alluvium (CGA) that is potentially liquefiable during strong earthquake shaking. The liquefaction potential of the CGA was evaluated using SPT-based methods. Validating analyses with the Loma Prieta event suggest the SPT-based liquefaction evaluation methods satisfactorily explain the lack of liquefaction observation during the event. The influence of embankment overburden on the liquefaction potential of foundation alluvium was investigated as part of the study. The findings indicate that the presence of the embankment likely had only minor effects on the liquefaction resistance of the CGA.

The building of a cross-sea bridge on deep, soft soil in the deep sea requires the thick, soft soil reinforcement in advance. To this end, centrifuge model tests on the composite pile foundation were conducted. In the tests, the pile lengths were 18 cm and 9 cm, respectively. The diameter was 8 mm, and the pile spacings were 3d, 5d and 8d, respectively. The test results indicate that the foundation settlement value with raft can be effectively reduced by the raft between stress balance layer and basement. Corresponding to pile length of 18 cm, the foundation settlement value with raft was 39.1% of the one without raft. When the pile length was 9 cm and the pile spacings were 3d, 5d and 8d respectively, foundation settlement values with raft were about 13.4%, 35.2%, and 52.9% of the settlement value without piles and raft. Settlement value with the pile spacing of 8d was almost the same with the raft foundation settlement without pile. Corresponding to the pile spacings of 3d, 5d, and 8d, when the pile length was 18 cm, the settlement values were 12.5%, 76.5%, and 90.1% of the one without piles, respectively. When the pile spacing was 8d, raft foundation settlement with pile length of 18 cm was 30% less than the settlement with pile length of 9 cm. According to the test results, relationship between settlement and pile spacing in two different conditions was based on whether the raft is involved. Through nine groups of centrifuge model test of foundation reinforcement with steel composite piles, settling characteristics and the effects of pile spacing, pile length and raft on foundation settlement were studied. The results of this study provide important reference for theory research and engineering applications of foundation reinforcement of extremely deep soft soil for sea-cross bridge.

In this study, the one-dimensional compression test was conducted to investigate the effect of zinc concentration on the compressibility of cement-solidified, zinc-contaminated kaolin clay. Post-yield compression index (C_c) and yield stress (σ_vy') of stabilized soils were obtained and correlated with initial zinc concentrations. It has been found that zinc concentration significantly affected the compressibility of stabilized kaolin clay in terms of C_c and σ_vv'. C_c decreased marginally and sharply when zinc concentration ≤ 0.1% and > 0.1%, respectively. σ_vv' decreased steadily with increased zinc concentration. To quantify the influence of zinc concentration on the cementation structure/bonding and, further, the compression characteristics of the cement-stabilized kaolin clay, the concepts of void index (I_v), Intrinsic Compression Line (ICL) and Sedimentation Compression Line (SCL) were used to analyze the compression test data. The locations of compression lines of the stabilized kaolin relative to those of ICL and SCL, were discussed. In addition, the convergence of compression lines of stabilized kaolin with ICL was analyzed as well. The results reveal that as zinc concentration increases, cement stabilized kaolin clay exhibits weakened cementation structure/bonding. It is demonstrated that zinc has a considerably destructive effect on the cementation structure/bonding, which in turn affects the compression characteristics of the cement-stabilized, zinc-contaminated kaolin clay.

The geotextile industry research and development is growing at a fast rate and new products of variable specifications are now available for earthworks and design engineers. The nature of the material as a filter and the engineering properties of the materials used made it possible to explore a wide range of application.The polypropylene is non-biodegradable and can stand aggressive chemical exposure when placed inside a soil media.This research is conducted on a polypropylene geotextile material locally manufactured in order to gauge the retention capacity for a range of opening sizes. Tests were carried out using an erosion plate subjected to controlled flow for a specific period of time. Erosion loss is computed for different products and different soil material. The results were viewed in relation to the soil grain size distribution and the geotextile material physical properties. The data provides helpful guide for the design of erosion control system using typical geotextile material.

Practice shows that the analysis of laterally loaded pile foundations is critically important. For years, numerous experimental, theoretical and numerical investigations have been conducted to predict the behavior of such systems. In this paper, we are describing a computational study to investigate large pile-ground system response under lateral load based on 3-dimensional (3D) OpenSees finite element modeling. Primarily, pile group behavior in nonlinear soil with and without a tension-cutoff logic is discussed. Distribution of load within the pile group is analyzed and the group interaction effects are examined. A robust and versatile framework (OpenSeesPL, http://cyclic.ucsd.edu/openseespl/) for computational analysis of pile-ground systems is employed to facilitate the pre- and post-processing phases. Further validation and calibration of the analysis framework may be conducted with the aid of case histories and experimental data.

This paper presents a theoretical study of effective soil density for small-strain shear wave propagation in saturated soil specimens in resonant column tests."Effective soil density" is a term used to consider the fraction of pore water that moves with the solid skeleton during shear wave propagation.The ratio of effective soil density to saturated soil density is always and is a function of porosity, specific gravity of solids, hydraulic conductivity, and shear wave frequency.In resonant column tests, the transition of effective density from the maximum value (i.e., saturated density) to the minimum value occurs primarily in coarse sands. In this region, the consideration of effective density is important in capturing the first resonance and interpreting resonant column test results.For soils less permeable than coarse sands (e.g., clays, silts, and fine sands), the consideration of effective density is not important and the saturated density can be used.For soils more permeable than coarse sands (e.g., gravels), the effective density is significantly less than the saturated density.

This paper reviews the geotechnical impact of the 2008 Wenchuan Earthquake gleaned from references obtained through internet and other sources. It briefly describes the seismo-tectonic and geological setting of the seismic source region, and discusses various geotechnical aspects of the event. Strong ground motions and fault displacements in a setting of steep mountainous terrains trigger many failures of natural slopes and man-made structures. Coseismic hazards include landslides, debris flows and formation of landslide dams. These ground failures not only increase earthquake damage and hamper rescue and recovery effort, but also generate a secondary flood hazard due to potential breach of landslide dams. The epicentral region has also many high water-storage embankment dams, which seem to fare quite well. Finally, the event has set the stage for latent geological hazard caused by earthquake-weakened slopes and/or metastable landslide/debris-flow deposits. This hazard would materialize either in an aftershock or heavy storm/snow-melt event.

Steel sheet piles have been widely used in construction of earth retaining structures and seepage barrier walls. It is known that steel is susceptible to corrosion when in contact with the environment. Therefore, durability of permanent steel sheet pile structures against corrosion needs to be evaluated as part of the design of these structures. This paper presents a case history where a corrosion protection system was designed and constructed for a 1.5-kilometer-long steel sheet pile barrier wall at a lake cleanup project site. The system includes a coal tar epoxy coating and a sacrificial anode cathodic protection. The paper discusses the methodology used to design the corrosion protection system and provides details of its construction. The case history is expected to be beneficial to geotechnical engineers working on similar projects.

The main purpose of this study was to use the discrete element method-based program, Particle Flow Code (PFC^2D), to study the mechanisms of particle movement (including contact forces, displacements, flow pattern, and the strength parameters between particles) of debris flow.The following simulated results based on the particles passing through the falling door tests and the inclined trough tests can be obtained:The particles start moving as the friction angle between particles equals 28-30 degrees as the trough slope is 22 degrees.The angle of repose is about 56-60 degrees for particles with friction angle of 28-30 degrees.The angle of repose is about 34-38 degrees if subtracting the trough slope.This angle of repose can be treated as the macro friction angle of the debris materials.The obtained results are similar to the friction angles, 34.7-36.2 degrees, tested by Su (1998) on valley deposits using direct shear test.It is found that slope angle is the major reason that controls the incident of debris flow, and then friction angle between particles, volume concentration of the mixture, and friction angle of a slope.The characteristics of the simulated debris flow from this study are very similar to the actual debris flow.

Hyperbolic models are frequently used in practice to represent nonlinear shear stress-shear strain behavior of soils in equivalent linear and nonlinear dynamic modeling problems. Several researchers have proposed shear modulus reduction (G/G_max versus \xg) curves that use the hyperbolic model as their basis, with parameters that fit the models to cyclic laboratory test results. However, cyclic laboratory tests often are not run to failure shear stress levels. Consequently, the model G/G_max curves are well constrained by the data at small-to-moderate shear strains, but do not necessarily provide an accurate representation of soil strength at large shear strains. In some cases, the shear strength can be grossly inaccurate, which may result in significant errors for analyses involving shear stress levels at or near failure. In this paper, a new hybrid methodology is presented that permits simultaneous matching of: 1) the conventional shear modulus reduction curves that are well calibrated at small-to-moderate shear strains and; 2) the soil shear strength at large strain. This hybrid approach produces shear modulus reduction curves that result in corresponding, hyperbolic-like, smooth, stress-strain, backbone curves. These curves assure that the soil shear strength consideration is accurately represented for the purposes of both equivalent-linear and nonlinear analyses.

This paper discusses some of the most critical issues related to behavior of clay particles during consolidation phenomena through radial flow using central vertical drains. The macro mechanics of consolidation of soil is evaluated using scanning electron microscopy of consolidated clay samples based on development and dissipation of excess pore water pressure using vertical drains viz. coir-jute fiber drain and sand drain of varying diameters. Micro compressibility of soft clay mass based on volumetric porosity is deduced with Quanti-image analyzer with system of micro-structure characterization (MIC) software, thereby introducing a relation between rate of consolidation, rate of suction and rate of repulsion of clay-pore water system. This methodology has given better clues to find eco-bio vertical geodrain for rapid consolidation of soft soils.

There have been many historical cases where pile foundations suffered severe damage from liquefaction during an earthquake. The pile foundation loading and deformation due to liquefaction-induced lateral spreading are two major factors evaluating the behavior of pile foundations for bridges in seismically active areas. This paper presents a preliminary seismic evaluation of pile foundations for a bridge located in the coastal area in California subjected to a lateral spreading hazard. Pile groups consisting of 44 cast-in-steel-shell (CISS) piles with a diameter of 6 feet are proposed to support two bridge towers located on each side of a channel. Liquefaction potential is considered high at the site under the design earthquake event. Liquefiable layers of thicknesses of 28 feet and 26 feet are expected at the east and west sides of the channel, respectively. The potential for lateral spreading under the design earthquake event towards the channel is considered to be high as the stratum of liquefaction appears consistently widespread at both sides of the channel. Pile group deformation and forces resulting from the potential lateral spreading were evaluated for both tower foundations by using an equivalent nonlinear static analysis procedure recommended by the California Department of Transportation (Caltrans). This approach imposes a lateral spreading soil displacement field on the foundation model in which p-y curves are used to consider nonlinear response of soils. Pile foundation deflection, moment and shear corresponding to the imposed lateral soil displacements were then calculated and compared with the performance criteria.

Thermosyphons are usually used to maintain the stability of construction foundation in cold regions. This paper presents a three-dimensional, finite element, thermal analysis conducted to simulate 30 years' running to study the cooling effect of the road subgrade with thermosyphons. The spacings of the thermosyphons used in the computation are 3 m, 4 m, 5 m and 6 m, respectively, and the diameter of the thermosyphon is 89 mm. Preliminary results indicate that the thermosyphon can lower the soil temperature substantially, and the cooling effect becomes less effective as spacing increases. For the geological conditions in the Tibetan Plateau, the effective operating radius of the thermosyphon with the diameter of 89 mm is about 1.6 m and the reasonable spacing between thermosyphons is 4 m.

In order to investigate water equilibrium of expansive soil subgrade, equilibrium water content of recompacted, expansive soil under different stress conditions in wide suction range is measured, and the key parameter for numerical analysis is obtained. A two-dimensional subgrade-atmosphere interaction model is developed and used to numerically simulate variation of water content of a covered, expansive soil subgrade in 10 years. Compaction control criterion for expansive soil subgrade is proposed considering the equilibrium water content. The results show that equilibrium water content is stress-dependent in suction of 0-3 MPa. Water content of subgrade increases slowly in 8 years and reaches equilibrium state finally. The optimum water content of wet compaction is close to equilibrium water content and natural water content of expansive soil in moist and hot climate. Compaction of expansive soil subgrade should be controlled in light of wet compaction criterion.

By conducting field dynamic tests on a ballastless track-subgrade system located in the Milestone JJK86+274 of Beijing-Tianjin Intercity Dedicated Passenger Line in China, vibration velocities and their distributions of the center and the edge of slab track, the concrete roadbed edge, and the subgrade surface were measured under excitation of a high-speed train at the highest speed of 344 km/h in the world. The relationship between the maximum vibration velocity of slab track and train speed was also obtained. The results of the dynamic tests expand findings of the behavior of the ballastless track system on high-speed railway in Germany. An S-shaped growth curve along with a fitting formula are derived and can be used to describe the correlation between the maximum vibration velocity V_d and train speed V. This paper provides a useful reference for evaluation, prediction and control on dynamic response of ballastless subgrade, high safety, high stability and high comfort for the running of high-speed trains in the future.

Piled-raft structure is used in high-speed railway to deal with the deep quaternary strata compressibility foundation, which can effectively control total settlement and differential settlement, aiming at the efficient, economical and fast treatment subgrade deformation. In this paper, based on the field test of Beijing-Shanghai high-speed railway project, the dynamic response of high-speed railway piled-raft structure is investigated. The results agree fairly well with those obtained by theoretical approach and show that settlement of the subgrade is controlled effectively by the piled-raft structure.

In order to study the unsaturated, remolded strength characteristics of loess, loess in a region of Gansu is selected for remolded samples. By using direct shear apparatus for unsaturated soils made by GDS of the UK, 12 unsaturated direct shear tests are made. The test results show that the stress-strain curve ofthe rebuilding unsaturated loess is strain hardening or weak strain softening. In addition, comparison of different matric suction under the conditions of the stress- strain curve shows that, along with the increase of matrix suction, unsaturated loess shear stress increases. In the same matric suction, shear stress increases gradually with increase of vertical load. In the analysis of the curve of cohesion C and matrix suction of unsaturated loess, the cohesion and matric suction of C was in low suction range linear growth relationship, so that the double variable strength formula is suitable for the loess in that region, and the friction angle related to the matric suction is φ_h =2.23 °.

Ground liquefaction remains as one of the major causes of infrastructure damages during earthquakes. The concept of artificially creating "desaturation or partial saturation" in liquefiable soils has been introduced to improve soil liquefaction resistance. This paper describes the theory and computational implementation to simulate the effects of desaturation on soil liquefaction resistance. The theoretical basis of the computational model is developed based on Biot's consolidation theory. Computational simulations were conducted to simulate the behaviors of soils under cyclic loading.From these, the trends of pore water pressure evolution and liquefaction resistance of soils under different initial degree of saturation were predicted. Simulation results show that as degree of saturation decreases from 100% to 86%, the number of loading cycles to initialize liquefaction increases from 86 cycles to 237 cycles or increases by 175%.This indicates that desaturation or artificial induction of partial saturation in soils is an effective approach for soil liquefaction mitigation.This can be accomplished by, for example, introducing air bubbles into the pore spaces of soils. While promising, the procedures to introduce air bubbles into soils need to be further developed in light of practice requirements.

When dam overtopping occurs, the hydraulic power of the impinging water jet might cause rock foundation erosion/scouring, which potentially could lead to significant consequences. As future flood events are expected to be more severe due to climate change, dam erodibility is becoming a significant risk concern. This paper: (1) provides an overview of erodibility analysis, including project examples to illustrate the use of Annandale's Erodibility Index method, Bolleart's Comprehensive Scour Model, and Goodman and Shi's rock block theory; (2) describes a reliability-based approach to address the issues of uncertainties, such as first-order reliability method and importance sampling; and (3) discusses the measurement of rock properties for developing probabilistic models, including a field example of using land-based LiDAR to obtain rock joint statistics.