To perform wind-resistant reliability analysis of structures，it is necessary to simulate the stationary stochastic fluctuating wind field． Simulating the stochastic fluctuating wind fields through spectral representation method ( SＲM) often involves thousands of random variables，leading to the“curse of dimensionality”． On this basis， this paper proposes a new method for points selection in high-dimensional random-variate space to ensure the accuracy and efficiency of SＲM． The new method mainly consists of three steps: the decomposition of high-dimensional random-variate space，the points selection in the orthogonal subspaces，and the random pairing to obtain the representative points and assigned probabilities in high-dimensional random-variate space． The proposed method is used to investigate the horizontal wind field acting along the height of a 30-story frame-shear wall structure． The results show that the proposed method is of high accuracy and efficiency for the simulation in the second-order statistical sense and a complete probability set can be constructed by only using a small number of representative sample processes． Besides，the proposed method can be further combined with the probability density evolution method to elaborately evaluate the probabilistic response of wind-induced vibration of structures．

The traditional level set method overcomes the defect that it cannot automatically create holes by arranging a certain number of holes in the initial design，but the distribution form of holes causes a strong dependency on the initial design． The improved parameterized level set method could automatically create holes，but the initial design still affected the optimization results． The level set band method can combine the level set method with the variable density method and introduce a level set band region near the zero level set，and the material density in this region was interpolated by the level set function． The intermediate density materials were gradually removed through gradually reducing the width of the level set band in the iterative process，so that the structure finally converged to a 0-1 solution． Introducing level set band can improve the continuity of objective function and constraint function in topology change by combining variable density method，so as to reduce the influence of initial design on optimization results． The compliance minimization problem proves that this method can effectively reduce the initial design dependency of the optimization results and get a better objective function value． Furthermore，the influence of the reduction speed of the width of the level set band on the optimization results was studied，and this method was extended to the optimization problem with irregular design domain and reasonable optimization results were obtained

In order to explore the in-plane tearing behaviors and strength characters of fabric membranes under combined shear and tensile stresses，experimental and mesoscale numerical studies were conducted on a typical warp-knitted fabric membrane． Its specific tearing behaviors and strength characters under a series of stress ratios and stress levels，reinforced film types and coupled stresses were obtained，and the tearing mechanisms and failure behaviors of the material affected by shear-tension coupling stresses were analyzed． Ｒesults show that the fabric membranes could produce three kinds of appearance in terms of the orientation of the tear: line-shaped，crossshaped，and Z-shaped opening，which is depended on the crack orientation and stress states imposed on them，and the corresponding damage directions of yarns could belong to either warp or weft mode，instead of mixed modes． The stress states in warp and weft directions show the synergistic effect to offer the resistance to tear propagation， which affects the tearing strength obviously and leads to significant differences of tearing strength curves for different crack parameters in the stress space，resulting in the occurrence of stress strengthening at the stress ratio 1∶ 1． The stress state combined with reinforced film can obviously interfere with the tearing damage modes and crack position of membrane material，and the reinforcing effect varies depending on the stress ratios． In addition，during the tearing process，there is a coupling relationship between the tensile and shear stresses，and the tensile stress effect is the main one，while the shear stress has no obvious effect on the crack propagation and evolution． The research findings could provide references for the localized crack arresting of fabric membrane materials and safety assessment of membrane structures．

Most parts of south China are located in subtropical region，where hygrothermal environment will reduce the performance of prestressed CFＲP reinforced ＲC structure． In order to explore the influence of hygrothermal environment on the fatigue performance of prestressed CFＲP reinforced ＲC structure，the failure criterions were established in the paper based on the two failure modes of main reinforcement fracture and debonding failure of CFＲPconcrete interface under the combined action of hygrothermal environment and fatigue load and the fatigue performance of prestressed CFＲP reinforced ＲC beams in hygrothermal environment was analyzed by numerical analysis． The results show that the fatigue performance of the reinforced beams will be reduced in hygrothermal environment， and the effect of increasing the prestress degree of CFＲP on the predicted life of the beams presents two circumstances． When the ＲC beams are damaged due to the failure of the main reinforcement，the increase of prestress degree can significantly increase the predicted life of the ＲC beams． However，when the beam is damaged due to the interface debonding，excessive prestress will reduce the life expectancy of ＲC beams．

In order to avoid the adverse effects of modal transition and unreasonable mode selection on the accuracy of damage assessment results，a seismic damage assessment method for spatial grid structures considering the contribution of multiple modes was proposed． Firstly，based on the idea of modal pushover analysis and the structural dynamic theory，the equivalent static displacement expression corresponding to the multi-modal contribution under unit seismic force was obtained． Then，according to the difference between the equivalent static displacement of original structure and damaged structure，the structural characterization parameter of the earthquake damage degree， namely，the overall damage parameter，was generated． Finally，the state of structural damage and performance degradation was assessed based on the magnitude and change of the overall damage parameters of the structure． The numerical analysis results of single-layer reticulated shell model show that the overall damage parameter of the structure calculated according to the method in this paper can more accurately assess the seismic damage degree of the space grid structure based on considering the multi-mode modal contributions． The obtained results are in good agreement with the comparison of structural characteristic response． In addition，the accuracy of the assessment results can be guaranteed in the case of severe damages．

Under the long-term geological action, the saturated permeability coefficient of soil shows natural strong spatial variability. In order to explore the influence of saturated permeability coefficient variation on the failure probability of slope, the Green-Ampt model was improved and the relationship among the depth of wetting front, the safety factor of slope and the duration of rainfall was determined. Then, Monte Carlo method was used to establish the analysis framework of variation of saturated permeability coefficient, and the  influence of saturated permeability coefficient variation on the failure probability of slope under different rainfall intensity was discussed. The results show that in the initial stage of rainfall, the smaller of the coefficient of variation of saturated permeability is, the smaller the failure probability of the slope is. With the increase of rainfall duration, the failure probability of the slope increases gradually, and with the increase of the coefficient of variation of saturated permeability, the failure probability of the slope becomes smaller. In the critical rainfall range, the failure probability of slope instability rises suddenly, and the critical rainfall range becomes concentrated with the increase of rainfall intensity.

Aiming at the problem of structural parameter identification and unknown input inversion under earthquake, this paper proposed a method of structural parameter identification and seismic input inversion using only partial absolute acceleration measurements as the observation. Firstly, based on transform space method and least squares method, structural parameters above the first story were identified directly. Then, the stiffness of the first story was identified based on the identified modal information. Finally, the dynamic  equation was transformed into the transform space in which the correlation model of seismic input at adjacent sampling time was established and a simplified input inversion algorithm is presented. Results of numerical simulation and shaking table tests on a two-story reinforced concrete frame show that the proposed method has high accuracy and robustness for linear systems, and for nonlinear systems, the inversion results reflect the low-frequency characteristics of seismic input which can be used on seismic performance deductive analysis for regional structures and buildings.

In order to develop an optimal method to identify the initial bending stiffness of a simply supported beam, the beam was segmented to directly measure or indirectly calculate the rotation angles at the segmentation cross-sections under a known static load based on the idea of virtual segmentation. The bending stiffness of various girder segments was inversely obtained by establishing a relationship between the rotation angle and bending stiffness of each girder segment. The correctness and validity of the proposed method was proved by a finite element analysis and the experiment data. The condition number of a matrix was applied to analyze the influence of the number of girder segments and the magnitudes, types, numbers, and positions of the acting forces on the stability of the identification equation set, and the influence of rotation angle measurement error on identification accuracy. The results show that the number of girder segments, number of forces, and positions of the forces were the key factors influencing the stability of the identification equation set; the acceptable rotation angle measurement noise increased with the decrease of the number of girder segments and the increase of sensor accuracy; under proper loading and test conditions, the proposed method shows high robustness in identification of initial bending stiffness of beams. The proposed method is simple and practical because it requires no complex inversion algorithm, support of a finite element model, or pre-measurement of parameters such as girder cross-section dimension, reinforcement, or mate-rial properties.

Compared with ordinary Portland cement, geopolymer has the advantages of environmental friendliness, fast hardening and high early strength, smaller shrinkage, and better bond behavior. To investigate the feasibility of using geopolymer mortar as anchorage adhesive, pull-out tests were conducted on post-installed steel bars anchored by geopolymer mortar. Influences of substrate type (concrete, granite), anchorage depth, rebar diameter, and loading mode (monotonic, repeated) on anchorage performance of geopolymer mortar were investigated. Geopolymer mortar was compared with usual cement mortar and commercial organic adhesive, and its bond strength was compared with that of inorganic adhesive reported in literature. The results show that the anchorage performance of geopolymer mortar is significantly better than that of cement mortar and is almost equivalent to that of commercial organic adhesive. For a ribbed rebar with a diameter (d) of 16mm, when using geopolymer mortar as adhesive in concrete and granite substrate, a minimum embedded depth of 12d required by code can already satisfy the ancho-rage quality requirements. Furthermore, it is also showed that the anchorage performance of geopolymer mortar has no significant degradation under repeated loading, at a repeated loading level not higher than 70% of the ultimate load. Thus, geopolymer mortar can be used as anchorage adhesive for post-installed steel bars in the concrete substrate and granite substrate in practice.

The finite element model of concrete steel wire skeleton sandwich wall was established by ANSYS/LS-DYNA. The difference of displacement response of concrete steel wire skeleton sandwich wall under the action of shock wave, fragment, and a combination of the two was analyzed. And the influence of different axial compressive strength of concrete, steel wire skeleton yield strength, polystyrene foam density, angle of the inclined plug wire and arrangement, the thickness of the concrete on both sides and the bubble, the proportion of explosive distance and height and fragment size on the displacement response of concrete steel wire skeleton sandwich wall was investigated. The comparative test results show that the modeling method is reliable. The displacement of the wall under the action of single shock wave is greater than that under the action of single fragment and the combined action of the two, but the combined action of the two causes more serious damage to the wall. With the increase of the axial compressive strength of concrete and the thickness of concrete on both sides, the anti-explosion performance of the structure is enhanced. The steel wire skeleton yield strength and the density of polystyrene foam have little influence on the anti-explosion performance of the structure. The different angle and arrangement of diagonal filaments have great influence on the anti-explosion performance of the structure. The explosive height has little effect on the wall damage. When the proportional distance is fixed, the larger the explosive amount is, the more serious the wall damage will be. From the angle of wall damage, the larger the fragment size, the more serious the damage.

To evaluate the effectiveness of different stiffening measures, the comparative study on the axial compression behavior of the stiffened and unstiffened concrete-filled thin-walled high-strength square steel tube stub co-lumns was carried out in this paper. Firstly, the axial compression experiments of the stiffened specimens and the reference unstiffened specimens were designed and carried out, and the influence of yield strength of steel tube and type of stiffening measures on the failure patterns, load versus axial displacement (strain) curves and the mechanical indexes, e.g. bearing capacity, composite elastic modulus and ductility index, of the specimens was investigated. The results show that the stiffening measures can alter the destruction process of the specimens and the failure patterns of steel tube and improve the mechanical indexes of the specimens. However, there are no stiffening mea-sures that can thoroughly enhance the mechanical indexes of the specimens. Secondly, by determining the reasonable constitutive model of steel and core concrete, the finite element model of the stiffened and unstiffened concrete-filled thin-walled high-strength square steel tube stub columns under axial compression was developed and verified through experiments. Eventually, the effect of major parameters on the axial compression behavior of the stiffened concrete-filled thin-walled high-strength square steel tube stub columns was further analyzed with the verified finite element model. The results show that, the demarcation points in different stages of load versus axial displacement curves and mechanical indexes of the stiffened composite stub columns are different, and the stiffening measure of diagonal binding ribs is more effective to improve bearing capacity and ductility, while the stiffening measure of double-row steel ribs is more effective to improve composite elastic modulus.

One of the main failure modes of steel structure is fatigue failure, which is often accompanied by sudden fracture of steel. At present, there are little research on fatigue properties of Q690D high strength steel. In order to obtain the fatigue characteristic curve of Q690D high-strength steel, the high-cycle fatigue test of Q690D high-strength steel was carried out in this study. The S-N (stress-life) fatigue curve of Q690D high-strength steel was fitted based on the test results and compared with the existing fatigue test curves of standard fatigue formula and other strength grades of steel. The test results show that the fatigue performance of Q690D high strength steel is much better than that of ordinary steel, and is larger than the calculated values of the American steel structure specification (ANSI/AISC 360-10) and the European steel structure specification (BS EN 1993-1-9). It indicates that the fatigue performance of Q690D high strength steel is superior, and the calculated values of the specification are conservative. The fatigue curve of steel is highly correlated with its yield strength and has strong discreteness.

In order to investigate the influence of the contact mechanism on the composite laminates with multiple delaminations, free vibration of composite laminates containing a through-the-width delamination was analyzed with a four-region exact model in which the contact mechanism at the interfaces of the delamination is considered. By examining the solutions from the exact model and adopting equivalent model based on equivalent stiffness, the delaminated region was translated into a perfect, delamination free laminate which has the same geometric size and the stacking sequence but a reduced stiffness. Numerical calculations were carried out to justify the accuracy and effectiveness of the equivalent model by comparing the natural frequencies given by the equivalent model with those given by the exact model and the finite element model. On this basis, the free vibration of composite laminates with multiple through-the-widths delamination was analyzed using the established equivalent model and the finite element model. It is found from the analysis that for laminates with multiple horizontal delamination, the larger the total length of delamination is, the smaller the natural frequency is; for composite laminates with multiple vertical dela-mination, the natural frequency of laminates with uniform distribution of delamination along the thickness is the smallest. Generally, the natural frequency decrease with the increasing number of delamination, but the natural frequency is affected by the relative depth between layers simultaneously. The results show that the natural frequencies of the delaminated laminates from the equivalent model are in good agreement with the ABAQUS finite element solutions, and the correctness of the equivalent theory is verified. It provides strong technical support for the correct evaluation of the residual bearing capacity of the delaminated composite structures in engineering fields.

In the socket connection, the bridge piers adopt prefabricated piers. After the bridge piers are inserted into the reserved holes of the cap, high-strength fast-setting grouting is poured into the periphery of the bridge pier insertion end to compact the reserved holes. The structural reliability of the socket connection mainly depends on the grouting and socket depth. Pseudo static tests and ANSYS numerical simulations were performed on the socket specimens and cast-in-place specimens. The effects of the grouts adhesive properties, the elastic modulus of the grout and the filling thickness on the overall mechanical properties were studied. The results show that,when the grout is bonded well, the socket specimen has the same level of thrust stiffness and ultimate bearing capacity as the cast-in-place specimen; after the cement grout fails, a new force transmission path occurs between the bridge pier and the cap. The socket specimen still has good thrust rigidity and horizontal bearing capacity, but the reinforcement of the socket test piece on the base plate of the socket needs to be strengthened accordingly to improve the bearing performance of the cap; the elastic modulus of the grout and the filling thickness have little effect on the ultimate bearing capacity of socket specimens,and the horizontal stiffness of the socket specimen is not influenced by the thickness of the grout, but it increases with the increases of the elastic modulus of the grout.

An explicit time-domain method(ETDM) based on analytical response of unit impulse was proposed  to simulate the random seismic analysis of artificial seismic waves and building structures with response spectrum consistent with “Code for Seismic Design of Buildings” (GB 50011—2010). Firstly, explicit time domain expressions based on analytical response of unit impulse for single and multiple degree of freedom dynamical systems were derived. Then, the explicit time domain expression for single degree of freedom system was applied to solve the stochastic dynamic equation of the ground motion. Through iteration, the artificial seismic wave of the specified intensity and the artificial seismic wave that is consistent with the standard response spectrum can be simulated efficiently by taking the advantage of equation in quickly solving the dynamic response. Then, the explicit time-domain expressions of the simulated seismic wave and the multi-degree-of-freedom dynamic system were applied to the random seismic simulation analysis of structures and verified by a numerical example. The study shows that, based on the existing ground motion model and the explicit time domain representation of the unit pulse, the seismic wave which is consistent with the standard response spectrum can be simulated efficiently, and ETDM based on unit pulse combined with mode superposition method has sufficient accuracy and high efficiency in seismic random simulation analysis of structures, thus can be used for seismic time-history analysis of building structures.

The accurate simulation on the wind field is an important prerequisite for the wind resistance research for the super high-rise buildings. In this research, the mesoscale WRF (Weather Research and Forecast) model, which can simulate high atmospheric circulations, and the CFD simulation, which possesses high spatial and temporal resolutions, were combined together to simulate a typical hilly area around the Shenzhen meteorological tower. Based on the WRF simulation loaded with the Four-Dimensional Data Assimilation (FDDA) technique and the CFD simulation for an equilibrium atmospheric boundary layer flow, the multi-scale wind field simulation analyses for this complex terrain were performed. The simulation result verified the applicability of the Analysis Nudging (AN) scheme and the Surface Analysis Nudging (SAN) scheme in WRF simulation, and it shows the accuracy of AN scheme is relatively higher. Compared with the separate mesoscale WRF simulation, the coupling multi-scale mo-deling results are much closer to the meteorological measurement data. The mesoscale WRF model can provide more accurate inlet boundary conditions for the small-scale CFD models and the downscale refined simulation can be rea-lized through the nesting operation in the mesh interface. Thus the simulation accuracy of the wind field can be improved by employing current multi-scale WRF-CFD simulation technique. This study indicates that the multi-scale simulation on the wind field in boundary layer based on the coupled WRF-CFD technique will be an effective way to improve the accuracy of the wind effect analysis for high-rise buildings.

In order to study the influence of convective heat transfer boundary conditions on the temperature field distribution of steel box girder, a numerical simulation approach based on the modified convective heat transfer coefficient was proposed. Firstly, the commercial software Fluent was employed to simulate the wind velocity field of the steel box girder. Secondly, the convective heat transfer coefficient of the steel box girder computed and obtained with the Fluent was modified, and the temperature field distribution of steel box girder under different working conditions was further comparatively studied through finite element software ANSYS. The computation results of wind field on the surface of steel box girder show that compared to the inlet wind velocity, the wind velocities near the exterior surfaces of the steel box girder all decrease in different degree. Specifically, the wind velocities near the upper plate of the leeward side fairing and the bottom plate are reduced by 95.5% and 17.7%, respectively. The results of temperature field of steel box girder indicate that after introducing the modified convective heat transfer coefficient, the temperature differences in the steel box girder increase both vertically and horizontally. The vertically temperature difference will further increase after taking the auxiliary facilities on the bridge deck into the numerical simulation. Bridge deck pavement has little influence on the vertical temperature difference of steel box girder except the delay of the occurrence time of maximum temperature difference by 1 hour. In addition, the diaphragm can reduce the vertical and horizontal temperature differences of the steel box girder, due to its superior thermal conductivity. Therefore, the calculation of the temperature field of steel box girder should not ignore the influence of wind field near the surface of steel box girder, auxiliary facilities of bridge deck and diaphragm plate on the temperature field of steel box girder.

Yield and failure properties on the deviatoric plane for geotechnical engineering materials under three-dimensional stress state are characterized by anisotropy, which can not be reflected by most classical strength criteria. This paper extended the egg-shaped function from the meridional plane to the  plane and proposed the egg-shaped strength criterion for the deviatoric plane. Only 2 parameters α2 and β are required, which represent the curvature of egg-shaped function and the ratio of triaxial tensile strength to triaxial compressive strength, respectively. First of all, the feasibility of the egg-shaped function as yield/failure envelopes of deviatoric plane was mathematically demonstrated. Then the validity of the egg-shaped criterion was then analyzed and discussed combined with the results of several true triaxial tests. The numerical results show that the egg-shaped function can well simulate the strength envelopes with different geometric types depend on the defining parameters. When β is well determined, the egg-shaped function can successfully reproduce the test results as well as some other g(θ) expressions including the SMP criterion. And as α2 approaches 0, the egg-shaped strength envelope tends to be consistent with the SMP criterion. When it approaches 1, the egg-shaped strength envelope tends to be consistent with that of the Mohr-Coulomb criterion.

This paper used the coupled euler-lagrangian (CEL) method to establish three models, including a single-pile model, a single-pile model with pre-drilled hole, and a double-piles model with an isolated pile, aiming to study the soil displacement distribution around the pile and to investigate the control effect of pre-drilled hole and isolated pile on soil squeezing deformation. Results show that during the penetration process of pipe pile, the horizontal displacement distribution of soil along the depth varies with the penetration depth of pile, and the maximum displacement occurs near the depth of the pile tip. Furthermore, the soil under the depth of 6~7d (d is the pile diameter) below pile tip is not affected by the pile penetration. When the pile penetration depth is larger than 8d, the uplift deformation of ground surface caused by the yield failure of the deep soil gradually disappear. After the pile is fully penetrated, the horizontal displacement decreases exponentially as the distance from the pile increases. In addition, as the depth of pre-drilled hole increases, the control effect of pre-drilled hole on soil squeezing deformation becomes more effective. The isolated pile can significantly reduce the horizontal displacement of soil on the back side of the isolated pile, and can transfer the position of soil depth where occures maximum horizontal displacement from near the pile tip to a depth of 0.5L (L is the pile length). The CEL approach is well suited to study the influences of pile penetration on the surrounding soil, and can be widely applied to the researches of geotechnical problems involving the large deformations, such as pile installation and spudcan penetration.

The combination of high-strength concrete and steel fibers with different sizes can improve the performance of concrete deep beam, such as bearing capacity and toughness. In this paper, shear tests were carried out on one reinforced high-strength concrete deep beam and four hybrid steel fiber reinforced high-strength concrete deep beams. Through changing the volume of hybrid steel fiber and the distribution of web reinforcement, the failure mode, bearing capacity, deformation capacity, and reinforcement strain of specimens were studied. The enhancement mechanism of hybrid steel fiber for specimens was analyzed. The study results show that because of the hybrid steel fiber, the bearing capacity and deformation capacity of specimens are significantly improved, and the stiffness and bearing capacity decrease slowly in the later stage. The bearing capacity of specimens increases with the increase of the volume of hybrid steel fiber and web reinforcement ratio. The deformation capacity of specimens increased with the increase in the volume of hybrid steel fiber, but it increases first and then slightly decreases with the increase of the web reinforcement ratio. The short steel fibers function is to reinforce the concrete base-body and inhibit the forming of cracks. The long steel fibers role is to delay the development of cracks, transmit and bear the shear force at the cracks, and relieved the tensile stress of steel bars. Two kinds of steel fibers work together in the loading process of specimens. In addition, the calculation method to predict the shear capacity of hybrid steel fiber reinforced high-strength concrete deep beams was revised, with the consideration of synergistic effect of the two types of steel fibers. The calculated results were in good agreement with the test results.

In order to study the fire resistance of the cross-laminated timber (CLT) wall made of domestic Japanese larch, the CLT plate normal temperature material performance test and the CLT wall fire resistance test under the standard heating curve were carried out. The performance test obtained the interlayer shear strength, rolling shear strength and compressive strength along the grain of the CLT sheet. The fire resistance test results show that the failure mode of the wall at room temperature is mainly buckling failure, and the ultimate load of the three-layer CLT wall is higher than that of the five-layer CLT wall. Under the same layer composition, the fire resistance of the CLT wall decreases with the increase of the load-holding ratio of the test piece. Under the same load-holding ratio, the fire resistance of the five-layer CLT wall is higher than that of the three-layer CLT wall. Moreover, the CLT wall did not fall off during the test. The numerical simulation with ABAQUS shows that the temperature field and displacement development trends obtained are basically consistent with the test results.