The relations between energy release and energy dissipation are discussed; on the basis of the relations, a criterion for structural failure of salt rock is proposed. In order to verify the viability of the criterion, on the one hand, mechanical properties of salt rock based on Mohr criterion, D-P criterion and energy principles under different confining pressures of 1 MPa, 5 MPa, 10 MPa and 15 MPa are investigated; on the other hand, the failure criterion based on energy principle is carried out to study the data of conventional triaxial compression from other documents; the rusults reveal that the criterion based on energy principle only needs conventional parameters of rock mechanics, has clear physico-mechanical properties, and small relative errors between tested and theoretical values; so that this criterion can be used as judgment of salt rock failure.

Rock samples with different water contents need prepare precisely in many rock mechanical tests; but current methods are not able to meet the accuracy. This paper proposes a method to precisely prepare rock samples with different water contents. Its principle is that dried or saturated rock samples, kept in the environment with a constant humidity, and they will absorb moisture to increase weight or lose moisture to decrease weight until a constant water content for rock samples is obtained. The key point in this procedure is to create an environment with a constant humidity. For this purpose, a method is proposed based on the vapor-liquid equilibrium theory. At the beginning stage, chemical potential of water component in rock samples does not equal to that of water component of vapor in humidistat environment, which leads to different vapor pressures, transformation of water components and variation of water content in rock samples. When vapor pressures reach an equilibrium state, a precise constant water content homogeneously distributed in whole rock sample is obtained. This method is capable to obtain the water contents of rock samples with very slight differences which can not be obtained by the current methods, and also to precisely prepare the soil samples with different water contents.

Based on the field test data of pavement roughness, the computational parameters of vehicle loads are derived, and they are used to calculate the dynamic responses of road structures. By regarding the road structures as a three-dimensional multi-layered system, the dynamic responses and the influence depth of the road subgrade induced by the moving vehicle load are studied. The governing equations of motion are solved by employing the Fourier transform technique; and the theoretical solutions of the vertical stress of subgrade are deduced with the exact stiffness matrix method. A simplified three-layer structure composed of pavement, subgrade and half-space ground is obtained by employing the method of Odemark equivalent hypothesis of thickness-modulus. Then, the distribution of vertical stress and the influence depth of road subgrade are derived. The influence of different parameters such as load amplitude, equivalent pavement thickness and the subgrade modulus on the influence depth are carefully discussed; and the formulas of subgrade influence depth induced by vehicle loads are further derived. The method developed herein may be used as a preliminary reference in road engineering design.

Mercury intrusion porosimetry (MIP) test is used to study the change of pore size distribution of expansive soil during wetting-drying cycle. It is found that the microstructural parameters of expansive soil are increasing with the increasing number of cycles, such as the total volume of the pores, the porosity and the average pore diameter. After theoretical analysis and contrast experiments, the cause of test result deviation mainly stemms from two respects: one is the volume shrinkage of expansive soil samples during the wetting-drying cycle; the second is the influence of pore shapes and limited test pressures. And a reasonable amendment is given to the test data. Furthermore, the soil-water characteristic curves (SWCC) of expansive soil during wetting-drying cycle are calculated through the modified pore size distribution curves which are obtained from MIP test. Finally, the change rule of matrix suction of expansive soil during wetting-drying cycle is studied to deduce the relationship between the matrix suction and number of cycles. The results show that the moisture content 28% is the critical point; when the moisture content is larger than the critical value, the matrix suction increases linearly with the increase of cycle number; when the moisture content is smaller than the critical value, those are on the contrary. The paper gives a new approach to explain the wetting-drying effect of expansive soil mechanical behaviors.

A number of integrated and pre-existing hole sandstone specimens under cyclic drying and wetting have been prepared and tested under compression with meso-mechanical testing system. The processes of crack initiation, propagation and failure behaviors of specimens are observed by microscope. After several cyclic times of drying and wetting, the cracks initiate at lower load level, and the initial cracks generally are tensile cracks. The crack propagation in integrated specimens is three-dimensional at random, but in the specimens with pre-existing hole, the cracks start to grow at the hole wall firstly. In specimens with pre-existing hole, shearing cracks initiate from the compressive stress concentrated areas of the hole wall under natural state; on the contrary, tensile cracks may initiate at the central zone of tensile stress under cyclic drying and wetting. Moreover, the cracking progress is more complicated and its rupture form is multiple. Compared with natural specimens, the specimens under cyclic drying and wetting become more ductile, and their strength parameters are lower .

To investigate feasibility of using dewatered sewage sludge as cover material of landfill in humid areas, an experimental program is carried out to measure the changes of strength and permeability of sludges collected from wastewater plants in Hangzhou, which are respectively soaked in de-aired water and leachate for one or two months prior to testing. Meanwhile, a parallel test is performed on a low plastic clay utilized as traditional cover material. The experiment results indicate that the unconfined compressive strength of dewatered sewage sludge is higher than that of raw sludge, and it decreases by 5.4% and 18.9% after two-month soaking in leachate and water respectively. Effective friction angle of dewatered sewage sludge increases by 7.5° after the soaking process, and its cohesion shows a decreasing trend. Permeability of sludge increases of 3-5 times after the soaking process. The experimental results demonstrate that shear strength of the dewatered sewage sludge meets the prescribed requirement as landfill covering material.

The post-peak deformational characteristics of rock materials play a leading role in safety of geotechnical engineering. So it is important to study the post-peak stress-strain relationship. Based on the evolution of post-peak strength parameters of rock material, a general method for establishing the stress-strain relationship is presented. Firstly, strength criterion is selected to establish the connection of stress with strength parameters. Then, the relationship between strength parameters and strain softening parameter is determined. Finally, the post-peak stress-strain relationship can be obtained by strength parameters which are as intermediate variables. On this basis, based on Mohr-Coulomb strength criterion, regarding maximum principal strain as strain softening parameter, on the assumption that the cohesion and friction angle are the piecewise linear functions of , the method of solving express of post-peak stress-strain relationship is obtained. In numerical cases, the stress-strain curves under different confining pressures are obtained and they agree well with the test data. It reveals that the method proposed by this paper is reasonable and effective in studying post-peak stress-strain relationship.

Based on the thermodynamic mixture theory model which can consider the mechanical interaction occurring on the interface between two components in unsaturated soils is proposed by Wei and Muraleetharan, an evolution equation of outflow in an unsaturated sample is developed. The equation describes well the variation of outflow with time in a multi-step outflow experiment. According to the evolution equation, the method that can predict hydraulic parameters of unsaturated soils, including the soil-water characteristic curve and the hydraulic function, by transient multi-step outflow experiment under a given series of matric suction, is built. Through the multi-step outflow experiment, it is seen that, for the low liquid limit silt and clay, the predicted values of saturation and hydraulic conductivity under the given matric sunction which are calculated from outflow data of the multi-step outflow experiment, are very close to the measured values of that measured by the combined laboratory system.

High fill rigid culverts have been widely used in trenches of mountain area over soft ground. However, the present theories can not address the soft ground bearing capacity under the high fill rigid culvert very well; so an extremely need of ground bearing capacity may involve in some improper ground treatments, which probably induces adverse effect on culverts. In this study, a finite element numerical simulation and lab tests are carried out to investigate the soft ground bearing capacity under high fill rigid culverts. The influence factors including foundation embedded depth, foundation width and consolidation of soft soil are discussed in detail. This research reveals that the characteristic values of ground bearing capacity approximately linearly increase with the embedded depth of culvert foundation as the foundation embedded coefficient is less than 5. However, the influence of embedded depth of culvert foundation on bearing capacity gradually decrease with increasing the value of N while N is larger than 5. The influence of culvert foundation width on the characteristic value of soft ground bearing capacity is tiny, that can be ignored in the design of practical engineering. Moreover, lab tests results show that consolidation degree and consolidation pressure have inter-affect on the cohesion and/or frictional angle of soft ground soil. The cohesion and/or frictional angle increase with the consolidation pressure; the increasing rates will be faster for higher consolidation degrees. The frictional angle of soft soil is more easily affected by consolidation degree and consolidation pressure than cohesion. The ultimate bearing capacity of soft ground under high fill rigid culvert increases with increasing the consolidation degree of soft soil; the total increment is generally more than 36% while the consolidation degree reaches to 90%. Furthermore, it also nonlinearly increases with increasing the consolidation pressure; however, the increment decreases with the increasing of the consolidation pressure.

In order to obtain the maximum fines content meeting frost heaving ratio and compaction effect simultaneously, the frost heaving properties of fine round gravel soils (FRGS) under different fines contents and dry densities are studied by means of standard Proctor compaction tests and frost heaving tests. The results show that the optimum fines content of compaction effect for FRGS samples is 9%. The freezing process of FRGS samples can be divided into four stages, namely rapid freezing area, transition area, sub stable area and stable area. The FRGS is weak frost heaving fillings when the fines content is below 10%. The maximum fines content is 9% which is meeting the frost heaving ratio and compaction effect simultaneously. The frost heaving ratio of FRGS samples increases firstly and then decreases with increasing of dry density; namely there is a most unfavorable dry density. After 48 h freezing, the frozen section water content of FRGS samples is larger than initial water content. The water content distribution curves present S-shape and then gradually present inverted triangle with increasing of fines content. The increasing of dry density effectively blocks the water migration path.

The unconsolidated and undrained triaxial test is carried out on the intact infilled joint soils exposed at some major hydropower station under high pressure whose confining pressure is between 5 and 30 MPa to simulate the undrained boundary condition generated by rapid construction in-situ, and the mechanical behaviour of the infilled joint soils is discussed with the variation of physical property for each sample. The experiment results and failure mode are mainly controlled by high confining pressure: all the stress-stain relationships belong to strain hardening type, besides the samples after failure are presented as waist drum, and plastic failure occurrs. Moreover, a large amount of particle crushing occurrs among the wet particles, it is considered as the prime reason for the trend line of strength envelope decreasing with the increased confining pressure. The test data show discreteness for the effect of variation of physical property, and the sensitivity analysis (partial correction coefficient) indicates that: confining pressure p and saturation degree have a major effect on the initial elastic modulus and failure stress ; the void ratio e and the particle size distribution ( ) have the minimal effect. In addition, the equation proposed according to the shear behaviour can be used to estimate the strength features ( and ) of the infilled joint soils preliminarily in practice.

Based on two types of sliding surface, arc and arbitrary curve, the calculation methods of slope stability, including Sweden method, simplified Bishop method, simplified Janbu method, corrected Janbu methods, Morgenster-Price method, Sarma method and unbalanced thrust force method, are studied. Research results of above-mentioned methods are summarized. At the same time, four forms in Morgenster-Price method inter-slice force function f(x) are chosen, i.e. the constants of 0.1, 0.5, 1.0 and semi-sine function, and the formula in unbalanced thrust force method is improved. Through comparing some examples and analyzing the characteristics of these methods with different slice numbers slope angles, slope heights, and the conclusions can be obtained as follows: the factor of safety (FOS) calculated by Swedish method and simplified Janbu method is the least, and the results obtained by simplified Bishop method and the unbalanced thrust force method are quite close to the corrected method. The FOS calculated by arbitrary curve sliding surface method is slightly smaller than that of circular sliding surface method; and the critical sliding surface got by this method is close to the critical circular sliding face; so it shows that circular sliding surface, as an approximation of the critical sliding surface, can meet the need of practical engineering. Adopting a small amount of slices can obtain high accuracy of FOS in homogeneous slope; and a certain number of slices are needed to ensure reliability of the results in heterogeneous slope. In Sarma method assumption for the normal force and shear force on the side of slices makes calculated FOS larger than that obtained by other methods in homogeneous slope; then there are unsafe considerations, but Sarma method can consider the impact of heterogeneity in slope on slice method. When the slope shape (such as slope angle, slope height) changes, difference of FOS calculated by corrected Janbu method under two types of sliding surface is small; but the difference of FOS calculated by other methods is slightly larger. In Morgenster-Price method inter-slice force function f(x) has little effect on the calculated results.

A series of one-dimensional and isotropic compression tests have been carried out on undisturbed and remolded samples of marine soft clay taken from Shanghai and Jiangsu, with different sampling methods. The compression curves, compression index , and swell index are obtained from the tests. According to the normalized compression curves of undisturbed and remolded samples, the effect on compression characteristics of the soil structure is confirmed. The compression curves after the yield stress are extended to determine a reference void ratio which is used to quantify the fabric simply. Based on the compression index obtained from oedometer tests, the correlation between the compression index and void ratio e10 or reference void ratio are obtained. From the compression index between oedometer and isotropic tests, it is shown that they are the same on the whole, so the relationship obtained from the oedometer tests can be used for isotropic test results. In terms of the same relationship between the compression index and the reference void ratio between undisturbed and remolded samples, the characteristics of yielding of structure are discussed. The conclusion that the consolidation pressure is larger than the structure yielding stress, most of the bonding is broken, the difference of compression characteristics is caused by the difference in the fabric, is obtained. The conclusions especially the relationship between the compression index and reference void ratio provide important reference value for engineering practice.

Based on the anisotropic K0 consolidated soil constitutive model, the soil around the cylindrical cavity is classified into elastic zone and plastic zone. According to the cylindrical cavity expansion theory and boundary conditions, the closed-form expressions for the total stresses, the plastic zone radius and the excess pore water pressure of the K0 consolidated saturated soils are obtained. Meanwhile, a comparative results between the anisotropic constitutive model and modified Cam-clay model indicate that the different initial consolidation conditions have great effect on the stresses and excess pore water pressure induced by the cylindrical cavity expansion; the stresses and excess pore water pressure calculated with the K0 consolidated soil constitutive model are obviously greater than those from the solutions based on the modified Cam-clay model; but the plastic zone radius shows a contrary tendency.

The influence of unloading magnitude and porosity on stress release characteristics of sandy soil arching is analyzed by laboratory tests. Meso-study is also performed by PFC2D, the further study is focused on effect of particle friction property, stress and displacement field variation and stress release section division during arching. The results show: (1) The results of unloading magnitude effect by PFC2D agree well with laboratory tests; it's feasible to study stress release characteristics of sandy soil by PFC2D. (2) The vertical stress release in stress release section is larger than that in stress increase section, while horizontal stress release is not; the lateral pressure coefficient increases as getting closer to the unloading site in stress release section while the regularity is reversed in stress increase section. (3) The friction property of sandy soil reduces the influence of arching on stress release and lateral pressure coefficient. (4) The stress release factor increases as unloading magnitude or porosity increases; the lateral pressure coefficient increases in stress release section while decreases in stress increase section as unloading magnitude or porosity increases.

Chlorite schist is a kind of soft rocks, and its large deformation and collapse due to excavation are greatly harmful to engineering safety. In order to investigate the mechanical property of chlorite schist, triaxial compression test is conducted and a new internal variable is defined considering the influence of stress state on plasticity evolution. According to test results, evolution characteristics of cohesion and friction angle are that the cohesion decreases firstly approximately linearly and then parabolically, but friction angle always decreases approximately parabolically. Meanwhile, based on test data, evolution of dilatancy angle is opposite to that of friction angle. Finally, mechanical model of chlorite schist is proposed considering hardening-softening and dilatancy characteristics; and then it is applied to simulate the triaxial test; the result shows a good agreement with the test data. Effort of this paper offers a rational mechanical model for engineering safety estimation and an important reference for the mechanical investigation of other similar soft rocks.

The dynamic strength characteristics of unsaturated silt have been studied by using a dynamic triaxial testing apparatus for unsaturated soils. Some test results have been obtained, including the characteristics of dynamic strength, liquefaction and pore-water pressure under the conditions of different saturation degrees, shear stress ratios and cell pressures respectively. The test results show that (1)the dynamic strength decreases with increasing of saturation degree; the relationship between the saturation degree and cycle number is linear in the semi-logarithmic coordinates at failure under the same shear ratio; (2)the liquefaction critical curve of unsaturated silt approximates to be the S-shape; the liquefaction resistance is decreasing with the increasing of saturation degree; the liquefaction is no longer taken place when the saturation degree is less than 60%; (3)the pore-water pressure increases in a way of 3-stage with the increase of the cycle number.

For the iron ore which is full of water and weak broken sections in complex environment, various methods, such as the thoretical analysis, laboratory research, numerical simulation and field experiment are firstly used to conduct the comprehensive research and to form a deep cognition about stability characteristics of the weak and gravelly ground of the iron ore and the key technology of tunnel construction; a whole set of construction technology that safely, rapidly and safely passing through the iron designing ore which is full of water and weak broken sections is formed. The key technology mainly includes reasonable cross-sections, scientific design of supporting structure, adopting active permanent support instead of passive two linings, selecting of optimum support parameters, the correct method of excavation, and informatization construction, etc. The application demonstrates that the key technology can solve the engineering problems effectively and ensure the safe construction; and it has significant advantage of economic benefit and social effect.

The energy analysis method for stability analysis of slope reinforced by prestressed anchor cables is presented based on geotechnical plastic limit analysis theory, in which the reinforcing effect of prestressed anchor cables is taken into account. The iteration interior point method and sequential quadratic programming method (SQP) are used to calculate the objective function of safety factor incorporating the strength reduction technique and the minimum energy principle of the instability state. Typical example calculation and analysis show that the location of prestressed anchor cable has significant influence on slope safety factor, the location of critical slip surface and the prestressed anchor cable length. Considering the influences of the location of prestressed anchor cable on slope safety factor and length of anchor, the optimal settings of single anchor cable should be located in the lower slope region. Slope reinforced by multi-prestressed-anchor-cables can guarantee the security reserve of slope, and can avoid the local stability problem of slope effectively.

Predicting roof settlement according to measured series is the main measure of roof stability analysis. Based on mathematical morphology (MM), a new method is proposed to improve predicting accuracy. Settlement series is treated as a group of signal which is decoupled to some components by using multi-scale mathematical morphological method. Least squares support vector machine (LSSVM) algorithm is used to predict each component. Rebuilding these components turns to be the final predicting result. According to the measured settlement time series on AIII monitor point of #762 stope pre-controlling roof in a gold mine, an MM-LSSVM predicting model is established to test the new method. The predicting results show that the components present more regularity and less interference than the original series. Comparing with the traditional predicting methods, the mean absolute percentage error (MAPE) and root mean square error (RMSE) of MM-LSSVM have a great decrease; and the predicting accuracy is improved obviously.

On the backgroung of the first-level soil excavation on the podium pit of Shanghai Tower which is constructed with top-down method, the effectiveness of earth berms on pit deformation is analyzed. Analytical results show that the results of calculation agree with field measurements for the lateral deformation of soil. In deep and large top-down excavations, the effects of the changes of retaining wall thickness range from 0.8 m to 1.6 m; and whether to consider lowered wall on the pit deformation are not significant when the first level soil is excavated. When the retaining wall is modeled with beam elements, the calculated value of wall lateral deformation is 10%-21% larger than that when the wall is modeled with solid elements; the reason is that the latter can consider the effect of resistance moment which is generated by the friction between the wall and the soil. The lateral anti-pressure F firstly decreases slowly with the decrease of the width of berms b. When b reaches about 20 m, F reaches the minimum, then gradually increases with the decrease of b. To the given excavation depth of berm boundary , there is a critical value for b; when b decrease to , F reaches peak; if b is further reduced, the wall deformation is increasing rapidly. And increases with the increase of . Finally, based on these results, the design method for earth berms in basin excavation is given.

In view of the deficiencies of grey relational analysis used in current ways for selecting secondary risk factors when evaluating debris flow risk, multiple correlation coefficient is used to select secondary risk factors. Thus, multiple factor system consisted of primary risk factors and secondary risk factors for evaluating debris flow risk is established. On this basis, a new independent information data fluctuation weighting method is used to calculate the weights of secondary risk factors. With the new weighting method, the relations among the primary risk factors and secondary risk factors can be decreased and the contribution of secondary risk factors to debris flow risk can be expressed. With the selected factors and weights, the calculation formula of debris flow risk is proposed. Then the new formula is taken to calculate the debris flow risk of 12 debris flow gullies in Dongchuan city. The results coincide with the added status of secondary factors which improves the rationality of the method.

The classical Prandtl mechanism could not reflect the effects of inserted depth of wall penetration well and has limitation to reflect the effects of the nonhomogeneity and anisotropy of the clay. An improved analytical formula based on a modified Prandtl mechanism is given, which obeys the upper bound theorem of plastic limit analysis. This formula could reflect the effects of inserted depth of wall penetration well, and the effects of the nonhomogeneity and anisotropy of the clay could also be considered. In the meantime, considering the effects of the nonhomogeneity and anisotropy of the clay, a shear strength reduction finite element method (SSRFEM) is developed through the secondary development of commercialized software. The reasonability and reliability of the proposed modified mechanism and SSRFEM are verified by compared with the multi-block upper bound method. And the formula which is verified by a case study can give proper safety factors considering the influence of inserted depth of wall penetration, nonhomogeneity and anisotropy. Related conclusions can bring references for excavation designs.

According to the Biot theory in saturated soil, a finite-infinite element method in frequency domain is put forward for analyzing elastic wave propagation. The finite element is utilized to model the near-field medium and the infinite element with shape functions based on analytical solutions is used on the artificial boundary, which can represent wave extending to infinity in far-field. By means of proposed method, a numerical example is presented to demonstrate the effectiveness and accuracy of the method. The parametric study is conducted herein in detail for screening effects of active isolation by open trench in saturated ground. The results show that the isolation effectiveness of open trench is mainly related to the depth of the open trench. Usually increasing the depth gives rise to improving isolation effectiveness. When the depth of the open trench is determined, the distance between the trench and vibration source should not exceed 1.5 times of shear wavelength of saturated ground. Furthermore, the width of open trench has little influence on the screening effectiveness relatively.

Landslide susceptibility assessment and prediction are the important contents of the first solution for the decision-making of prevention and mitigation of disaster. Starting from the geological environmental conditions within Longzi county in Tibet, the 13 influencing factors are selected, including elevation, slope and so on. According to landslide risk classification standards established, landslide risk rating is divided into high, moderate and low. Based on combination weighting-unascertained measure theory coupled evaluation model, 20 landslides? risks in study area are assessed and analyzed by unascertained measure function, combination weighting and confidence criterion. Each landslide risk is obtained. Compared with evaluation results of fuzzy matter-element and catastrophe theory, the evaluation results are consistent with actual situation and reasonable. Therefore, the model can provide a new idea for landslide risk prediction.

Based on load transfer method and Winkler model, an elastoplastic solution of tunnel-pile interaction is given by the two-stage method to analyze the vertical behavior of adjacent single pile due to tunneling. The long-term and the short-term vertical behaviors of adjacent single pile are analyzed. The results are compared with that obtained from some existing centrifuge model tests, and then a good agreement is observed. The influences of the distance between tunnel center line and the ground surface, the distance between tunnel center line and the pile axis, ground loss, pile length, pile diameter and pile strength on the vertical behavior of adjacent single pile due to tunneling are analyzed; and the influence regularity is presented. The results show that: (a) the distance between tunnel center line and the ground surface increases, the pile settlement and pile axis force firstly increases and then gradually decreases; (b) the distance between tunnel center line and the pile axis increases, the pile settlement decreases gradually, the pile axis force increases to a certain value and then gradually decreases; (c) with the increase of ground loss , the pile settlement is continuously increasing, the pile axis force gradually increases to a stable value.

Evaluation of seismic stability of rock slopes and the calculation of starting velocity of earthquake-induced landslide are always difficulties for prevention of seismic failure of rock slopes. Combined with the large-scale high-speed landslides induced by 5.12 Wenchuan earthquake, taking Tangjiashan high-speed landslide for example, the kinetic failure mechanism of consequent bedding rock slopes under strong earthquake is proposed. The kinetic failure mechanism includes four stages: tension fracture– wedge splitting–sliding–shear breaking. And the formation mechanism of tension rupture, the leverage of wedge splitting rock masses and the its rolling friction effects of clastic stone, are described in detail. Using the elasticity theory, the failure criterion of locked–up segment of consequent bedding rock slope under earthquake is deduced. And the calculating formula of deformation energy of locked–up rock masses, while the locked-up segment is sharply cut off under earthquake, is also deduced. Considering the advantage releasing direction of locked-up rock masses deformation energy, the energy conversion principle is applied to propose a more reasonable and accurate burst starting velocity formula of the consequent bedding rock landslide, further determining the departure velocity of landslide. The formula provides an effective quantitative calculating method on the hazard scope and impact damage extent of landslide.

Based on a large amount of microseismic monitoring data and hundreds of rockbursts with different intensities occurring in the deep diversion and drainage tunnels of JinpingⅡhydropower station, relativity between rockburst and microseismic activity zone is studied. The conclusions are drawn as follows. (1) Microseismic activity mainly ranges from 3 times the tunnel diameter behind working face to 1.5 times ahead; and rockburst high-risk zone lies in the region as large as 3 times the tunnel diameter behind working face. Therefore rockburst high-risk zone is greatly in accordance with the main microseismic activity zone. (2) Potential rockburst high-risk zones in tunnel project are highly localized in the region already excavated as large as 3 times the tunnel diameter behind working face and the region under construction as large as 1.5 times the tunnel diameter ahead of working face. (3) Seismic events and rockburst distribution are characterized by regional aggregation. Some of them happen in the aggregation region of seismic events. Others happen on the edge of aggregation region of seismic events. That is an inherent phenomenon in the failure progress of rock mass and closely related to local stress concentration on the edge of aggregation region of seismic events.

How to choose a reasonable footage of tunneling cycle in complicated conditions has been a problem in tunnel construction. Based on the existing research results, a spatial relational expression about tunnel excavation is proposed, and a calculation method which based on the on-site displacement monitoring data is also work out. Combined with on-site construction conditions and monitoring data about Erlang fault-rupture zone at #9 work area of Guanjiao tunnel, the spatial effect expression of displacement during tunneling is established, results indicate that the spatial relational expression can reflect the displacement change with the changes of time and space. On the basis, the methods of spatial effect analysis and footage optimization of tunneling cycle about tunnel excavation in the soft rock with fault-rupture zone are achieved, and the footage of tunneling cycle in several large fault-rupture zones is optimized. Practice has proved that the method is feasible. Advantage and application scope of the method are analyzed too.

The research of deepwater mooring systems and foundations has become a frontal and critical topic in the field of the international ocean engineering. Drag anchors are increasingly used, and it plays a key role in deepwater mooring systems because of better performances both in pullout capacity and deepwater installation. During the drag anchor installation, a complicated interaction happens between the anchor and the installation cable at the shackle. In order to analyze the interaction between the drag anchor and the installation cable effectively, it can be equivalently analyzed through investigating the drag force at the shackle. Studying the calculation methods for the interactional force between the anchor and the installation cable is significant for understanding the forces acting on the anchor in seabed soils and selecting the tonnage of anchor handling vessels. Herein, the expressions of the drag force at the shackle are firstly derived based on the mechanical models for the drag anchor and embedded cable both for cohesive and noncohesive soils. Besides, experimental data are employed to examine the drag force predicted by the developed methods. It is concluded that both the theoretical expressions can reasonably predict the drag force at the shackle. However, the prediction results based on the mechanical model of drag anchors are more accurate and stable. The precision and stability of the two calculation methods for the drag force have been compared. Detailed parametric analysis is also performed for exploring the parametric effects on the prediction models of the drag force at the shackle. Comparing with experimental data, reasonable values of the key parameters in the prediction model are suggested, so as to provide important reference value for analyzing the soil-structure interactional problems.

Strong nonlinear and implicit performance functions are normally encountered when geotechnical engineering is complicated. An artificial neural network (ANN)-based four order moments method, allowing to make full use of statistical information of basic random variables, is proposed to address this issue. In this method, the relation between the basic random variables and responses is established to approximate the implicit performance function; then the values of the implicit performance function in the mean point as well as partial derivatives can be obtained for the four order moments method. The first four order moments are approximated by expanding the implicit performance function into a second-order Taylor series; the corresponding higher-order moments can be estimated via Pearson curves. In restraint of the assigned moments of the implicit performance function, the probability density function of the performance function can be expressed by applying the maximum entropy theory. Subsequently, the reliability results are easily worked out. From the comparison between the results of numerical example and practical engineering calculated by different methods, it is shown that the proposed method is valuable and applicable to reliability analysis of complex geotechnical engineering involving implicit performance functions.

In order to adequately master settlement calculation empirical coefficient of CFG pile composite foundation under embankment load, an analysis is taken out based on foundation information, subgrade design parameters and settlement field test data of 24 sections in Wuhan-Guangzhou high-speed railway. Empirical coefficient of settlement calculation’s variation and time effect features of settlement after construction are studied. The conclusions are as follows. Under embankment load, penetrating high-strength CFG pile composite foundation has outstanding deformation resistance, and its corresponding empirical coefficient changes between 0.63 and 0.17, which is lower than recommended value in Technical Code for Ground Treatment of Railway Engineering by 30 percent; foundation settlement shows a significant time effect features; and ratio of settlement after construction to final settlement decreases with time; the ratio of penetrating high-strength CFG pile composite foundation and time before stable of foundation settlement increases obviously with faster embankment filling rate and larger pile spacing.

Contact treatment of the numerical manifold method (NMM) mainly includes contact detection, including point-point and point-line contact, open-close iteration for determining the correct constrained state, and contact transfer by which effective contact couples can be transferred to the next time step. The whole process is slightly complicated and inefficient. Besides, the contact modes adopted in the NMM are quite different from the real physical contact state. The no binary search (NBS) contact detection algorithm proposed by Munjiza maps all elements into regular cells and connects the elements by the list structure effectively. The contact detections are limited within the cell and the four neighboring cells. With low requirement of random access memory, contact detection operation is just proportional to the number of the elements, and the efficiency is accordingly increased significantly. With the defined ‘potential’, contact pressure, belonging to the distribution category, is calculated by the contact overlapped area, which is closer to the real contact state, and avoids the cumbersome process involved in the NMM. As a result, it is introduced to the NMM as an alternative contact treatment method. The effectiveness is verified by the examples.

Strain rate has important influence on mechanical properties and failure of the fractured rock mass. The influences of medium strain rate on stress-strain, crack tip stress, characteristic stress and damage of rock mass, fracture expansion of crack of sandstone with a single fissure are researched under uniaxial compression using two-dimensional particle flow code (PFC2D). The stress-strain curves of joint rock mass present obviously fluctuation and stress mutation. The stress mutation index is defined in order to quantitatively analyze the degree of stress mutation. The stress mutation fluctuation is more acute with the increase of strain rate; and after peak strength it is stronger than before. The fluctuation amplitude of stress is increased before peak strength but reduced after peak strength with the increase of the fissure angle. The fracture stress of crack is enhanced with the increase of strain rate; and the tangential shear stress is increased on the whole; but the normal stress obviously reduced with the increase of fissure angle. The loading stress when crack tip is fractured, the critical dilatancy stress σd and the peak strength are all increased with the increase of strain rate. The dilatancy of rock mass can be immediately triggered by the fracture of crack tip; and and are generally more close to when strain rate is lower. The more the cracks of damage and macroscopic fracture are, the stronger the degree of damage and fracture of rock mass are with the increase of strain rate; and especially the end effect of specimen is more significant. The initial fracture is begun in the way of I type wing crack; its length is related to the strain rate and the dip angle of fissure.

In order to automate and simplify the process of complex geological object modeling, based on analyzing the spatial characteristics of complex geological layer, a new automatic modeling method which aims at the geological layer including faults is put forward. This method firstly preprocesses the geological drilling material and geologic report figure to get the data of discrete points; the data will be stored in a database. Then choosing a modeling method based on surface model, the top surface and bottom surface models of the geological entity are established separately in the process that originates from borehole stratified information to profile map and ends up at geologic surface. According to the topological space, the method constructs faults with their fault lines and insets them into the original surface. Finally, by using contour line joining to generate the side-surface, it sutures those surfaces into a closed 3D geological object model. A 3D visualization system which is developed independently is used to verify this method, and it is successfully put into use to construct the 3D geological modeling of a coal mine in Shanxi Province. Experiment shows that the method is fast and efficient with high automation; the effect of the constructed model that expresses complex geological structure is good.

Energy storing and dissipation are the underlying mechanisms of the macromechanical responses of granular materials subjected to shear failure, while they are difficult to measure in laboratory. We implemented a user-defined contact model considering rolling resistance to the commercial software PFC2D, and made a calculable method to count the energy components based on the first law of thermodynamics. Then the energy storing and dissipation through the whole sample are investigated in a series of numerical biaxial compression tests by discrete element method (DEM). Four kinds of friction are adopted, i.e. sliding and rolling (S-R), sliding and non-rolling (S-NR), non-sliding and rolling (NS-R) and non-sliding and non-rolling (NS-NR). The results show that the energy is mainly dissipated in the type of sliding rather than rolling. And at a small biaxial strain, the input energy is mainly stored as elastic energy with a small portion dissipated and the dissipated energy is globally distributed through the whole sample. While a large biaxial strain is achieved, the dissipated energy gradually turns dominant and the majority of energy dissipated in a narrow zone, showing the significant localization. In addition, the main direction in which the energy is dissipated shows strong anisotropy at all loading stages.

On the basis of using the models of stress corrosion and pressure solution established by Yasuhara et al, the effect of fracture aperture closure on the fracture stiffnesses is considered. Aiming at a hypothetical model for geological disposal of nuclear waste in an unsaturated dual-porosity-fractured rock mass from which there is a nuclide leak, two computational conditions are designed: (1) the stiffness coefficients of a fracture are linear functions of magnitude of fracture aperture closure; (2) the stiffness coefficients of a fracture are constants, then the corresponding two-dimensional numerical simulations for the coupled thermo-hydro-mechanical-migratory processes are carried out by using finite element method; and the states of temperatures, rates and magnitudes of fracture aperture closure, pore and fracture pressures, flow velocities of underground water, nuclide concentrations, fracture stiffnesses and normal stresses in the rock masses are investigated. The results show: there are no obvious differences between the temperature fields, pore and fracture pressures, flow velocities of underground water, nuclide concentrations in the rock masses of two cases; fracture aperture closure is produced mainly by stress corrosion; the magnitudes of fracture aperture closure in two cases are closer each other at the same computation time; and those of case 1 are slightly larger; in the rock mass of case 1, the nearer the vitrified waste is, the higher the fracture stiffnesses are; and the larger stress values and higher stress concentrations occur in the vicinity of the vitrified waste.

A process of continuum excavation has been simulated using discontinuous deformation analysis (DDA) which is introduced element birth and death method. Comparing the calculated result with finite element method, we prove that DDA method is effective in cavern’s simulation. Based on the geological exploration data of Dagangshan hydropower station, DDA models of the underground cavern group with different combinations of joints are established. We analyze the failure modes of underground cavern group’s excavation, and explain the failure mechanism briefly. Based on Monte Carlo method and excavation design of Dagangshan hydropower station, random distribution of joints is simulated; and developments of displacement and stress are researched in the sub-step excavation process. Studies show that: displacements of all the blocks around the cavern are pointing to the cavern when excavation is completed; deformations of tailrace surge chamber and underground main powerhouse’s sidewall are relatively large; major principal stresses of the blocks around the cavern are parallel to the caverns’ boundary line; stress of the tailrace surge chamber relieves obviously; and blocks of sidewalls have slid; the three caverns interact on each other in the excavation process; the effects of tailrace surge chamber and underground main powerhouse on the main-transformed cavern are more obvious. The study shows specific parts which are need to be reinforced in the excavation process; and it provides a basis for optimal support measure of the project.

According to the conventional rockburst criteria, it is easy to determine the approximate area of rockburst. However, the damaged shape of rockburst (DSR) can not be computed based on these theories. This paper puts forward a method of simulating rockburst. The first stage of predicting DSR can be studied through critical stress. Then the second stage can be done by introducing pre-existed crack. Finally the scope of rockburst can be got through extended finite element method (XFEM). This problem can be overcome relying on the secondary development of ABAQUS. Through a model of fracture mechanics, it is easy to verify the feasibility and accuracy of XFEM. The DSR of Galongla can be got through simulation technology. And the result can provide guidance for the construction of underground project to prevent rockburst.

It is difficult for modelling 3D strata if different stratigraphic sequences exist among the boreholes. The problem can be solved through following method: deducing the inferior tables of stratigraphic sequences from the main table of stratigraphic sequences which is provided by the survey data with the information of the boreholes, have different stratigraphic sequences from the main table, is the first step. The main table and the other tables become a group of tables of stratigraphic sequences. The relationship that a borehole belongs to which table of the group can find out by using spatial probability analysis and the corresponding relationships between boreholes and tables then are set up. The problems of connecting the boreholes which have different stratigraphic sequences change into the problem of connecting the different tables. A technique that can be used to automatically figure out the corresponding relationship between two tables by searching for maximum count of same stratum is put forward. With this technique, all the tables in the group can be combined into a final table. Basing on it, the algorithm for building layer model is set up. At last, the validation of this algorithm is proved by using it in modelling 3D strata of the slope at Zhu-Yong expressway section K120+030~K120+300.

For solving stability problems of soil-mass under plane strain condition, finite element upper bound solution based on linear programming needs to linearize second-order cone constraint, forming by Mohr-Coulomb (M-C) yield criterion. The commonly used method is to replace the circular region projected from cone by external polygon. In order to improve accuracy of linearization, the direct way is to increase external polygon edges. This will result in the decision variables of linear programming model containing a large number of plastic multiplier variables, and leading to increase difficulties or even infeasible of calculating process. In order to solve the problem, a second-order cone linearizing method proposed by Ben-Tal and Nemirovsky is introduced and has been embedded into the coded finite element upper bound program. It is found that the results obtained by Ben-Tal method and results obtained from external linear polygon method are confirming each other after case study. Moreover, Ben-Tal method can improve the accuracy of linearization of M-C yield criterion effectively by appropriately increasing the number of decision variables and equality constraints. Simultaneously, it has proved that the formation of linear programming is small-scale. It is expected to be applied commonly to finite element upper bound solution based on linear programming model.

There is little domestic comprehensive information to introduce the development of triaxial test machine for more than 30 years. In order to do better analysis, design and manufacture of rock triaxial test machine for domestic universities, research institutes and manufacturers, the domestic conventional and true triaxial test machines are summarized and introduced on the basis of collecting a lot of reference, reviewing the history of the test machine development and doing thorough investigation and analysis. And the rock test machine developments of the creep, static and dynamic, high-low temperature and shear are introduced. Different performances triaxial test machines and their parameters of a few representative domestic users are enumerated. The future development trends of rock test machine which are turning into CT scanning, triaxial shear seepage, triaxial simulation test systems and in-situ true triaxial test system are pointed out. The results can provide a reference to the development of triaxial test machine and offer help to users who will buy a new test machine.