Natural clays have their structure which must be destroyed during the deformation process . In this paper a new kind of constitutive model capable to describe the damage of soil structure has been proposed. A structured clay in this model is regarded as an assembly of lumps of different sizes, and it is assumed that the total deformation consists of 3 parts, i.e. elastic deformation of lumps, plastic deformation due to sliding between lumps and irrecoverable deformation due to crushing of lumps. Usually yield function is introduced to describe the plastic deformation, however, a new damage function is proposed here to describe the additional deformation due to the damage of soil structure. Finally, a stress-strain relationship expressed in incremental form is formulated and a procedure is suggested to determine the model constants.

Pipe jacking in soft soils may induce ground movements, including ground settlement and axial movement of soils below the ground surface. These movements can lead to damages of nearby structures and utilities . Both theoretical analyses and field observations have shown that pipe jacking in such a kind of soft soil ground as in Shanghai area, movements of soils around the pipes is of three-dimension. Based on the basic principle of semi-analytical method, a procedure for the computation of soil movements is developed which discretize axial direction while adopting displacement functions in radial and circular directions . Regular functions are set up and semi-analytical element method is constituted. Displacement functions, stiffness matrices and load matrices are presented in details. By making use of the semi-analytical element method , the three-dimensional soil movements problem of pipe jacking is transfered to a one-dimensional numerical calculation. The value of a calculated case of pipe jacking in soft soils is compared to that of observations. The results show that semi-analytical element method is a satisfied way to predict soil displacements induced by pipe jacking in shallow soft soils. Due to the sharp reduction of the number of the element, the time needed for processing this problem is much less. Some useful conclusions have been obtained according to the analyses and calculations.

By means of proper numerical modeling, four kinds of pile composite foundation ( from the most flexible pile to the stiffest pile)with different cushions are tested numerically in this paper. The strengthening mechanism of the cushion in pile foundation is studied and discussed further.The optimized design and construction method of the cushion is then proposed for group pile composite foundation . The numerical tests show that ( 1) the bearing capacity of the pile composite foundation is obviously increased by proper usage of cushion , because the cushion reduces the stress concentration greatly; ( 2)the proper thickness of the cushion of optimized within 0. 2 ～ 0 . 5 m through enough numerical tests from the basic consideration of the optimization of load sharing ratio between pile and soil .

This paper introduces two three-moduli incremental non-linear stress-strain relationships under triaxial stress condition . The relationship using K , G, J is useful for finite element implementation. Based on hypoelasticity , this relationshiop is generalized into a three-dimensional stressstrain relationship. Both tensor form and matrix form are given . The physical meaning of the typical items in the matrix form is explaned .

The behavior of stress, strain, pore pressure and static undrained shear strength of undisturbed or remolded marine silty clay samples under undrained cyclic loading conditions is analyzed. The degradation of undrained shear strength is related to the cyclic strain and accumulated pore pressure produced by undrained clyclic loading. Based on the principle of quasi-overconsolidation and the results of comparing the cyclic triaxial test on normal consolidated samples and the static triaxial test on overconsolidated samples, the relation between normalized undrained shear strength and quasi-overconsolidation ratio is estabalished. In order to estimate the degradation of undrained shear strength of marine sily clay , the cyclic triaxial test results of undisturbed and remolded marine silty clay samples are needed.

Considering the strain weakness behavior of fault and the visco-elasticity of surrounding rocks . A simple mechanical model for fault earthquake is put forward . First, the equivalency between the model and cusp catastrophe is proven . Then, the cusp catastrophe theory is applied to analyse the instability mechanism of the fault earthquake. It is explicated that the quick transferring of energy and the violent releasing of energy dont take place unless the enough energy from external world is input into the mechanical system and the stiffness ratio is less than unit . The period of seismogeny is both related to the period of loading owing to distant field displacement and to the visco-elasticity behavior of surrounding rocks.

Utilizing the scanning electron microscope, X-ray diffractometer, type 9310 microporosity analyzer, the microstructure and chemical composition of mudston, and its macro-variation of physical and mechanical character are measured and analysed , the mechanism of mudstone degradation and softening in water is systematically studied in this paper.

The analysis of the formation of shear band will benefit both the study of the engineering behavior of soils and the evaluation of their stability . Based on elasto-plastic model proposed by Sekiguichi and Ohta, the influence on shear band formation of the rotation of principal stress directions is investigated. It can be concluded that the directions along which local shear deformation may occur most probably depend on the rotation angle of the principal stresses, but the rotation of principal axis doesn’t contribute significantly to the instability by localization of the deformation .

The time series of displacement measured in-situ of landslide is regarded as the unstable random process in the paper. The short-term abnormalities are extracted by means of the homomorphic model and the identification criterion of abnormalities and computing formula of order degree are given. The critical-sliding abnormalities are extracted with one degree difference equation for high pass filter. The prediction is conducted with the grey catastrophe model through the abrupt date set composed of abnormality times .

In this paper, the structure is simulated as a uniform shear beam supported on a rigid foundation embedded in the soil represented by a uniform visco-elastic half space. The closed form structural responses are deduced using substructure method for seismic excitation represented by vertically incident SH waves. Control of structure is realized by setting TMD( Tuned Mass Damper)which makes the upward propagating waves absorbed at the top of the structure and no downward propagating waves reflected. Subsequently, the frequency-dependent character of the parameters of TMD is acquired . Furthermore, control of the first mode of the soil-structure system is realized by choosing the parameter of TMD. Structural responses and effectiveness of the control are compared and discussed for three working patterns named no control , completely control and first mode control .

In this paper, the method of reliability analysis on stability design of cement-soil retaining structure is established, where the indices of shear strength are regarded as basic variables. Based on 21 geotechnical data of Shanghai soft soil, JC-method is used to analyze and check the reliability indices and discuss the sensitivity of basic variables related to failure modes .

This paper performs an experimental study on disturbed ground during tunneling through static cone penetration test. Based on the analysis of soil mechanics parameters collected from the test , general influence laws of disturbed ground in shield tunnel are obtained, and quantitative description of disturbed ground is also discussed.

Mechanical model for pile-driving is presented according to the soil deformation behavior and yielding criterion of structural soft soils. The distribution of excess pore pressure is studied using the cavity expansion theory.

Detailed engineering geological investigations have been completed on Lingquan Temple Grotto, in Henan province. The Grotto that was constructed about 1, 400 years ago and consist mainly of hard compact crystalline limestone has produced serious geological diseases . A classification of the geological diseases is carried out in the present paper, they are: rock mass instability , water percolation and corrosion , and weathering diseases . Moreover, the reasons for the diseases are analyzed . The safety factor of the rock mass and the slope have been obtained respectively through analytical method and finite element method by assuming the seismic loading being pseudostatic. Finally, suggestions are given for renovating the Grotto.

Through tests on rock block similation of simularity of rock material and finite element calculation, the characteristics of friction and squeezing at main roof block corner are studied . It is found that the friction angle of main roof block corner is the residual friction angle, and the coefficient of friction is determined as 0. 5. Weak planes have an obvious and regular effect on the squeezing strength of main roof block corner, and the coefficient of roof block corner is determined as 0. 4 .

This paper applied the theory of fracture mechanics to quantitatively analyse the influence of the length , direction and internal of cracks on the rock strength . And the formulae for the initial rupture strength of the rock mass under hydraulic pressures are also derivated in detail . The result has verified that the formulae proposed in this paper is highly reliable in the assessment of the stability of engineering rockmass and rock slope with water inbiltrated due to torrential rain .

Original variables are greatlty correlated with each other in reliability analysis of geotechnical engineering . Several methods of HasoferLind reliability index calculation are stated , including the orthogonal transformation method , the improved first-order -second-moment method, the Monte-Carlo simulation method, and the optimal method . When using the method that directly uses optimal tool to solve reliability index β , the orthogonal transformation about correlation variables and the partial derivatives of the limit state function to the original variables are avoided, and Microsoft Excel 97 Solver is used to do the optimization. Two examples are calculated for the comparison of the methods .

The filter paper method for measuring the suction in soils is described . The homemade quantitative filter paper is calibrated with a pressure membrane apparatus, the suction calibration curves during dewatering and dewatering-absorting are obtained. The properties of these curves are compared with those of foreign filter paper, some conclusions have been summarized.

On the basis of discussing and analysing the current situation of reinforcing depth and affecting factors of the dynamic consolidation method for pile engineering, the present paper establishes the formula with unified dimenssions that determines the consolidation, expressed as . The application of the formula is also disccused。

Drawing the experiences of mechanism of reinforcing clay-soil foundation in the sand drain method, the present paper puts forward a method in which the saturated silty clay foundation is reinforced by dynamical sand-soil piles. Loading testing research on proving zone of a practical project has showed that not only the sand piles have high density and strength, but also have physical and mechanical properties of ground soil been significantly improved. Bearing capacity of composite ground has been greatly enhanced. Application of the method to the project also proves the testing results .

In order to calculate the thermal conduction coefficient of earths , neural network technique is used to map the non-linearity among the physical indexes of earth and an expected result is obtained. It is shown that a BP-network is useful to map the non-linear relationship among the coefficient and other indexes and this technique is more accurate than linear regression and further applications are expected.

According to the mechanics model of soils around pile while static driving pile, the cause of additional settlement is approached from different angles in this paper. Relevant calculation formulae are also presented. In addition, an example is calculated by them and the results are proved to be reliable and reasonable compared with the practical case history.

The constitutive model of interface with thickness and elastic-plastic constitutive model have been employed in order to consider the behavior of interface element and the plasticity of soils . The research has been made on how the pile geometry , mechanical behavior of pile, soil mechanical behavior and interface behavior have influence on the pile settlement by 2D-FEM . Subsequently , several conclusions have been drawn .

Based on a great number of in-situ tests, the paper presents the relationship of soil resistance vers depth and lateral displacement, the relationship of lateral foundation reaction factor vers soil indexes. Based on the principle of similitude, we have replaced non-linear FEM programming calculation with tabulation method - NL method. Case history of 39 piles has illustrated that the method is of reliable precision and practicability.

A procedure of multi-grid solution to adaptive nonlinear FEM is described in this paper. Numerical analyses on underground cavern excavatio n have shown the effectiveness of the procedure and the advantages o f adaptive multi-grid FEM in the simulation o f excavation process.

According to the principle of generalized plastic mechanics, the mathematical expression of the shear yield surface in the Lode angle θe direction is presented initially and a three-yield-surface model for soilmasses is established based on the experiment of previons researchers. In combination with the numerical example, many calculating schemes are designed. The comparison and analysis of various calculated results have shown the validity of the model. The effects caused by various factors on the deformation of soilmasses are discussed.

This paper presents a practical model for oil reservoir in field recovery. Based on stress-seepage coupled variational principles, it is proved, using the perturbation method, that the three-dimensional problem can be reducible to a modified plane-strain problem with variable thickness of the oil formation if the thickness varies slightly in the horizontal direction. The solution then obtained is the zero-order perturbation solution. The global finite element equations are given for the modified plane-strain problem. An example for secondary oil-recovery in hydraulic fracturing is analyzed numerically and the results a represented.

On the basis of the measured data from model simulation tests on rock-socketed pile groups subject to torsion, the variation and distribution laws of the measured forces in pile top, torsional moment against displacement and torsional moment against rotation are analyzed. The testing results are coincident with those obtained from rigid loading stand assumption.

The computing program of DDA is expanded and developed based on DDA computation model in this paper. Large displacements of slope and surface deformation due to underground excavation are discussed using the DDA model with results compared with those from DEM . It is shown that the results are in accordance with those o f DEM . All of the analysis proves that DDA can be effectively applied to large rock engineering.

By means of the least square method, the initial stress field parameters are identified according to the measured data of transformation of the rock when the underground tunnel is excavated. Considering the stochastic characteristic, a stochastic method to identify the initial stress parameters in rock mass is established in this paper. The affection of the stochastic characteristic of the transformation measurement is studied as well. The numerical results show that the lower the sensibility of the objective function to the identified parameters, the larger the standard deviation of the identified results, and the worse the identifiable characteristic of the parameter.