Abstract:In order to improve the crack resistance of wet joints in the negative bending moment area of ordinary concrete continuous beams and simplify the construction process，a new type of UHPC wet joint scheme for the negative bending moment area of concrete beam bridges was proposed. Taking a common concrete continuous beam bridge with a span of 30m as the background，according to the French UHPC structural design regulations and Chinese Specification for Design of Highway Reinforced Concrete and Prestressed Concrete bridge and Culverts（JTG 3362—2018），the reinforcement design of the bridge under normal service limit state is carried out. Based on the results of reinforcement design，the 1 ∶ 2 scale model test of UHPC new wet joint structure was carried out. The test results show that the crack resistance and load-bearing capacity of the UHPC new wet joint scheme without reinforcement welding and prestressed strands in the negative bending moment area are all in line with the engineering requirements. The test values are well fitted with the numerical simulation values. The introduction of UHPC can effectively limit the crack width of normal concrete（NC），significantly increase the rigidity of NC section，and reduce the redistribution of internal force and vertical deflection of the continuous concrete beam in the mid-span. Compared with the traditional wet joint structure without negative bending moment prestressed tendon，the new UHPC joint structure can reduce the redistribution coefficient of the internal force and vertical deflection of the continuous concrete beam to about 30%~50%. The parameter analysis shows that： for the concrete continuous beam bridge with UHPC new wet joint structure in the negative bending moment area，the length of UHPC along the longitudinal direction of the bridge should be taken to be 0.27 times the calculated span. The diameter of the longitudinal tensile main bars in the negative bending moment area can be unified to 20 mm，and the thickness of the UHPC layer can be 60 mm.

Abstract:In order to study the amplitude effects on self-excited aerodynamic forces of the bridge main deck section，a thin plate section was taken as the research object，and the amplitude effects of aerodynamic self-excited forces were studied by using computational fluid dynamics（CFD） method. First，the numerical simulation method of forced vibration was used to investigate the flutter derivatives and aerodynamic hysteresis phases of the thin plate section under different amplitudes. The aerodynamic spectrum characteristics of the thin plate section were analyzed. Then，the numerical simulation method of free vibration was used to study the flutter response evolution characteristics of the thin plate section. The numerical simulation results of forced vibration for the thin plate section show that the torsional amplitude has significant effects on the flutter derivatives of the thin plate section. At high reduced wind speed，the flutter derivative A*2 changes from negative to positive as the amplitude increases. The effect of vertical amplitude on the flutter derivatives of the thin plate section is relatively small. As the torsional amplitude increases，the sinusoidal value of aerodynamic hysteresis phase of the thin plate section changes significantly. When the torsional amplitude is larger than eight degrees，there are obvious high-order harmonic components in the aerodynamic forces of the thin plate section，mainly for the third and fifth order components. The high-order components of aerodynamic forces of the thin plate section caused by the vertical amplitude are not obvious. The numerical simulation results of the free vibration of the thin plate section show that the instantaneous frequency，damping ratio and the phase difference between the vertical displacement and torsional displacement change with the vibration amplitude during the process of flutter divergence.

Abstract:In order to provide a simple and effective method for the fatigue performance analysis of bridge cable steel wires，according to the microstructure of the steel wire and statistical analysis results of relevant test data，a prediction model of steel wire fatigue crack growth rate is established，and the calculation method of model parameters is given. The fatigue crack growth rate in the stable growth region is obtained on the original steel wire by the variable load indentation method. The prediction model can better describe the data obtained from the test and the fatigue crack growth law of the steel wire in the near threshold region in the literature. On this basis，the fatigue life of new steel wire and corroded steel wire is calculated by the fracture mechanics method and compared with the test data. The results show that the corrosion will not affect the fatigue crack growth rate of the steel wire. The mechanical properties of the uncorroded steel wire can still be predicted by this method. The change of fatigue life of the corroded steel wire is attributed to the change of the initial flaw size. For new steel wire and slightly corroded steel wire，the equivalent initial flaw size method can be used to calculate the fatigue life；when the steel wire is corroded seriously，the total fatigue life is basically composed of the crack propagation life，and the initial flaw size is the true depth of rust pit. In view of the randomness of pit depth distribution，the maximum pit depth should be used to calculate the fatigue life of corroded steel wire.

Abstract:The past studies primarily performed the deterministic dynamic analysis of vessel-bridge collision for a given collision event，which hardly reflected the contingency and probability characteristics of ship collision，as well as the damage evolution of bridge under impacts with various energy. For this reason，this study systematically investigated the vulnerability of a bridge under collisions of two typical vessels，where the residual capacity of bridge piers after vessel impacts was used as the damage assessment index. Firstly，a direct finite element （FE） simulation method was established to predict the residual capacity of axially-loaded RC column after lateral impact. The rationality of the simulation method was verified by the test results. Then，based on a typical continuous girder bridge，two different simplified FE model were established，compared and validated. By combining the validated simplified FE model and the response surface method，a vulnerability analysis method of the vessel-bridge collision was established，and the vulnerability curves of the bridge were obtained for the impacts of two kinds of typical ships. The results show that the response surface has good accuracy and is able to replace the complex nonlinear FE calculation；The response characteristics of the residual capacity of the pier are quite different for the impact of two types of vessels：the residual capacity of the pier under the impact of the bulbous-bow ship decreases uniformly with the increase of the ship speed，while the residual capacity is closely related to the critical ship speed under the impact of the barge and exhibits the bilinear characteristics so that the sample design needs to be segmented based on the critical speed；Under the same ship speed and quality，the probabilities of structural damage and failure caused by barge impacts are generally higher than those caused by bulbous-bow ship impacts，which should be paid special attention in the practical design.

Abstract:To investigate the effects of central buckles on the flutter stability of long-span suspension bridges，the Aizhai Bridge in China was selected as the engineering background. Based on a refined spatial-truss-girder model and according to the principle of stiffness equivalence of the main girder in all directions，an equivalent-single-girder finite element model was firstly established by using the displacement method of a cantilever beam. Subsequently，four different connection options between the main cable and the girder near the mid-span position，namely，a short suspender，one pair of flexible central buckles，three pairs of flexible central buckles and one pair of rigid central buckles，were considered and their effects on the dynamic characteristics of long span suspension bridges were studied. Then，based on the flutter derivatives obtained from wind tunnel tests，the time domain formulations of self-excited forces in the girder section，expressed in terms of convolution integrals of impulse response functions，were obtained using a nonlinear least square fitting method. Based on APDL （ANSYS Parametric Design Language） offered by ANSYS，a time-domain flutter analysis was realized. Finally，the influences of the central buckles on the critical flutter velocity，flutter frequency，and three-dimensional flutter states of the bridge were investigated. The results show that the central buckles can significantly increase the frequency of the longitudinal floating mode of the bridge and have greater influence on the frequencies of asymmetric lateral bending mode and asymmetric torsion mode than that of symmetric ones. The rigid central buckle can largely increase the frequency of asymmetric torsion mode. The central buckles have negligible impact on the critical flutter velocity because the flutter mode shape of the Aizhai Bridge is coupled with the symmetric vertical bending mode shape and the symmetric torsion mode shape. However，it has a certain impact on the flutter frequency and the three-dimensional flutter states of the bridge，which benefits the flutter stability. In addition，it is found that the phenomenon of complex beat vibration （called intermittent flutter phenomenon） appeared when the structural damping was very low，because the flutter frequency falls in an area where the natural frequency distribution is very dense.

Abstract:A method of identifying damping coefficient matrix in exponential non-viscous damping systems based on the modal test is proposed in this paper. By solving a constrained optimization problem，the optimal modified damping coefficient matrix satisfying the system characteristic equation is obtained；and considering the incompleteness of the modal parameters obtained in the actual mode test，the damping coefficient matrix of the system can be accurately identified by using the limited low-order modes. However，with the gradual enhancement of the non-viscous characteristics of the system，the number of modes needed to accurately identify the damping coefficient matrix gradually increases. Furthermore，a method of updating the imaginary part of the complex mode，which can be greatly affected by noise in the modal test，is proposed to satisfy the characteristic equation of the exponentially damping system in this paper；and as a result，the damping coefficient matrix can be accurately identified by using the updated complex-modal imaginary part. Due to the lack of research on the recognition of the relaxation parameter，an identification method of relaxation parameter which is independent of the imaginary part of complex mode is proposed based on the modal updating method. And then，through numerical studies，the applicability and effectiveness of the damping identification method and the complex modal correction method for the exponentially non-viscous damping model are verified，and the accuracy of relaxation factor recognition method is proved. Finally，the modal parameters of a cantilever beam are identified through the vibration test，and the applicability and rationality of the exponential damping model are discussed.

Abstract:Seven RC bridge decks that have been used for nearly 60 years were tested on flexural behavior after being strengthened with steel plates and composite steel plates and CFRP. The flexural capacity，stiffness，cracks and strain，as well as the failure modes of the test slabs under different strengthening methods were compared and analyzed. The results show that both steel-bonded reinforcement and composite reinforcement of steel plate and CFRP can effectively improve the flexural behavior of the test slabs. The flexural capacity of the specimens with 2 mm，4 mm，and 6 mm thick steel plates is increased by 52.5%，126.0% and 162.5%，respectively. Correspondingly，the flexural capacity of the specimens with composite strengthened slabs is increased by 87.0%，148.0% and 158.5%，respectively. When the steel plates and CFRP were combined to strengthen the existing damaged flexural members，the strengthening effect of CFRP can be brought into full play. The current codes and regulations for steel-bonded reinforcement can be used for the calculation of flexural capacity of existing damaged RC slabs. A calculation method for flexural capacity of existing damaged members strengthened by steel plates and composite steel plates and CFRP considering damage effect is proposed，and the calculated results agree well with the experimental results.

Abstract:In order to study the interface bond behavior of high strength concrete filled circle steel tube （HCST） after elevated temperatures and water cooling，22 specimens were designed to be subjected to static push-out tests，and the effects of concrete strength，maximum temperature，bonded length，constant temperature and cooling mode were mainly considered. The interface failure mechanism of the HCST after elevated temperatures and water cooling was revealed through experiments，the influence of various parameters on the bond properties was analyzed，and the formulas for calculating the ultimate bond strength and residual bond strength of HSST after elevated temperatures and water cooling were put forward. The results show that the load-slip curves of the loading end and the free end of the specimen subjected to elevated temperatures and water cooling are basically similar，and they can be divided into three typical curves. The longitudinal stress-strain distribution on the outer surface of the steel tube in the test is exponential. After elevated temperatures and water cooling，the bond strength of specimens changes little with the increase of concrete strength，and is inversely proportional to the bonded length. It is basically stable after the constant temperature reaching 60 minutes. With the increase of the maximum temperature，the ultimate bond strength first increases，then decreases，and finally increases，and the residual bond strength first increases and then decreases. Compared with the natural cooling specimens，the ultimate bond strength，residual bond strength and shear bond stiffness of the specimens subjected to water cooling are smaller，and the interface energy dissipation capacity is larger. The results calculated by the calculation formula of the bond strength proposed in this paper agree well with the experimental values.

Abstract:In order to study the influence of steel pipe distribution on the seismic performance of steel pipe multi- ribbed insulation composite shear walls，four assembled steel tube multi-ribbed thermal insulation composite shear walls with a scale ratio of 1 ∶ 2 were designed and manufactured. The failure modes and modes，bearing capacity，hysteretic characteristics，skeleton curve，stiffness degradation，deformation and energy dissipation of the walls were studied and analyzed through the axial compression test of one wall and the low-cycle reciprocating load test of three walls. The test results show that the compressive capacity of multi-ribbed composite shear wall under vertical load is significantly increased due to the arrangement of steel tubes in the rib column；for multi-ribbed composite shear wall under low cyclic load，the main failure mode is shear failure，and the failure is basically in accordance with the order of “filling block-rib-frame column”，which is compound with common steel ribs. Compared with the composite wall，the shear capacity of the steel tube multi-ribbed insulation composite wall is increased by 112%，and it has good assembled multi-ribbed composite slab，which provides a theoretical basis for the application of multi-ribbed deformation capacity and energy dissipation performance. The experimental study perfects the structure system of composite slab structure in high-rise residential buildings.

Abstract:In order to investigate the influence of steel fiber content on fiber distribution and orientation in steel fiber reinforced self-compacting concrete，four kinds of concretes were prepared using four different fiber volume contents. The relationship between rheological properties of fresh concrete and fiber volume content was investigated and the fiber density in the cut planes of beam specimens was analyzed. The flow of fresh concrete with four different fiber volume contents was simulated by ANSYS CFX，where the effect of fiber volume content on the rheological properties was considered. Based on the velocity field of concrete obtained by simulation，the fiber is simplified as a number of particles that are rigidly connected and the motion of fibers is calculated by solving the rigid body dynamics equation. Through comparison between the experimental results，it is found that the proposed simulation method can determine the distribution and orientation of fiber immersed in the concrete. Furthermore，the simulation results show that the segregation degree of fiber in the vertical direction of beam decreases with the increase of fiber volume content. On the contrary，the orientation angle between fiber and beam axis increases.

Abstract:To study the wind pressure characteristics and flow mechanism of square columns due to flow interference，taking the tandem square columns as the research object，large eddy simulation method is applied under the condition that Reynolds number Re = 8 × 104 and spacing ratio P/B=1.1~5. The influences of spacing ration on wind pressure coefficient，aerodynamic coefficient，non-Gaussian feature of wind pressure，and wind pressure correlation of the columns are discussed. The relationship between flow pattern and non-Gaussian characteristics of wind pressure is revealed. The results show that three major regimes are distinguished along with various spacing ratios，namely single bluff-body regime，shear layer reattachment regime and co-shedding regime，which are closely related with the specific wind pressure patterns. In the single bluff-body regime，the wind pressure near the recirculation zone between two columns shows significant non-Gaussian feature，whose correlation of wind pressure is strong. The shear layer reattachment regime，however，shows a weak vortex shedding in near wake，accompanied by a small wind pressure correlation but a large region of non-Gaussian wind pressure. Finally，in the co-shedding regime，due to vortex shedding of the upstream cylinder，the wind pressure on the side surface of two columns has a strong correlation and there is a large region of non-Gaussian wind pressure on the side and leeward surfaces of the downstream cylinder.

Abstract:In order to determine a reasonable value of basic wind speed at both sides of Taiwan Strait，comparative study of extreme and basic wind speed under different climate conditions are carried out in this paper. The measured wind speed data are obtained from a meteorological station in Taipei from 1961 to 2015. The extreme and basic wind speed are estimated through Gumbel distribution adopted by building wind resistant design standards at both sides of the Taiwan Strait. The annual maximum wind speed sampling method and typhoon wind speed sampling method correspond to the mixed and typhoon climate conditions，respectively. The results show that，for the identical wind speed observation data，the extreme wind speed obtained by building wind resistant design standard in Taiwan TB2015 is slightly less than that obtained by current national building wind resistant design standard（GB50009） in Fujian，and the basic wind speed is just opposite，but the estimated wind speed difference between the two standards is very small. The extreme/basic wind speed under typhoon climate is larger than the corresponding value under mixed climate，and is closest to the measured extreme/basic wind speed within an error less than 0.5%. For both sides of the Taiwan Strait severely affected by typhoons，the estimated wind speed under typhoon climate condition is suggested to adopt as the basic wind speed.

Abstract:Based on high-frequency force balance wind tunnel tests，the reason for the interference effects on rectangular high-rise buildings in actual engineering was investigated. On this basis，the aerodynamic interference effects of two rectangular high-rise buildings in different spatial positions were studied. The results show that the interference amplification effect of the wind loads on principal rectangular buildings is mainly caused by another rectangular high-rise building located at the side and back of the principal building with orthogonal arrangement，and the interference effect of interfering rectangular building in the downstream region of principal building is significantly higher than that in the upstream region. By increasing the spacing ratio of the two rectangular high-rises along the side and rear of the principal building，the wind load interference effect tends to decrease overall，and the acceleration interference effect first increases and then decreases. Furthermore，the interference range and intensity of the interfering building moving on the side of the principal building are higher than that moving behind the disturbed building. The maximum interference factor （IF） of the downwind shape coefficient of principal building is 1.41. The interference effect also significantly increases the lateral average wind load of principal building. The lateral interference factor （IF） normalized by the along-wind shape coefficient of single building can be up to 1.08. After further considering the dynamic amplification effect，the IFs of the base moment of principal building at the along-wind and across-wind direction are 1.49 and 2.28，respectively，and the maximum acceleration IF of the building is 1.23.

Abstract:Based on the assumptions of one-dimensional nonlinear consolidation of soil proposed by Davis and Raymond， the one-dimensional nonlinear consolidation problem of double-layered soil under constant loading is investigated by introducing the continuous drainage boundary condition. The analytical solution for the one-dimensional nonlinear consolidation of doubled-layered soil is derived by means of variable substitution method and separation of variables method. The rationality of the present solution is also verified by comparing with Xie’s solution. Based on the present solution， the effect of different interface parameter and nonlinear parameter on consolidation behavior of soil is analyzed. The results show that， under the continuous drainage boundary condition， the solution of the average consolidation degree，Us，defined as the settlement，is always larger than that of the average consolidation degree， Up， defined as the pore pressure， and the difference between Us and Up increases with the increase of Nσ（the ratio of final effective pressure to initial effective pressure）. In the continuous drainage boundary condition， the Us increases with the increase of Nσ， while the Up decreases with the increase of Nσ. Compared with the Xie’s solution， the influence of Nσ value on Up is smaller in the continuous drainage boundary. In addition， the soil interface parameters have a great influence on soil consolidation.

Abstract:This paper analyzed the constitutive relationship of soil-structure interface，studied the deformation characteristics and stress paths of interface under different normal stresses with laboratory direct shear test，and established a mathematical model reflecting the constitutive relationship of soil-structure interface. The test results show that normal stress and interface roughness are the main factors affecting the shear characteristics of the interface. The shear failure mode of the interface changes due to the existence of consolidation stress. Its value is related to the roughness of the interface and the shear properties of the soil. Moreover，we proposed an improved statistical damaging model considering the whole shear process of the contact surface，based on ignoring the thickness of the interface and combining with the “three-stage” mode of the shear process. The correctness of this improved model is verified by experimental data. Moreover，this model can reflect the global deformation around the soil-structure interface under certain normal stress.

Abstract:To explore the effect of excavation on the horizontal displacement of adjacent tunnels in soft soil area，firstly，the deformation mechanism of horizontal displacement of adjacent tunnels in soft soil area due to excavation is analyzed. Secondly，project cases on excavation of adjacent tunnels in soft soil area are collected. The importance of the factors affecting the maximum horizontal displacement of the tunnel is ranked by the random forest algorithm. Through analyzing the influencing factors statistically，a semi-empirical formula，which is relatively simple and convenient for engineering practitioners，is proposed. It can be directly used to predict the maximum horizontal displacement of the adjacent tunnel under excavation. The accuracy and applicability of the proposed empirical formula is verified by comparing with the measured data of the actual engineering cases published in the collected literature. Parameter analysis shows that: with the increase of excavation geometry，the maximum horizontal displacement of the tunnel increases in an approximate logarithmic relation，but the increasing speed gradually slows down. The horizontal displacement of tunnel is greatly affected by the horizontal displacement of retaining structure. It is approximate linearly correlated with the horizontal displacement of retaining structure. The distance between the excavation and the tunnel has a negative correlation with the maximum tunnel horizontal displacement. Furthermore，when the distance is less than twice the excavation depth，the influence of excavation on the tunnel is greater. Based on the proposed formula，the effect range of excavation on the tunnel is divided into zones，and the results can provide certain theoretical guidance for similar projects.

Abstract:In view of the limitations of traditional entry-type excavation stability assessment methods，this study explores the uses of novel data-driven machine learning methods to establish the excavation stability prediction models based on the random forest（RF） and K-nearest neighbor（KNN） methods. The proposed methods are based on 399 case histories from eight Canadian mines，covering a wide range of rock mass rating（RMR） and span，with stable，potential unstable and unstable cases categorized into ternary and binary groups. A ten-fold cross-validation method is applied to optimize the hyper-parameters during modeling. These two machine learning methods can capture the complex relationship between the excavation stability with RMR value and span without any assumptions of the underlying relationship. The results indicate that the accuracy of the binary classification results are slightly better than the ternary prediction results. For the binary classification circumstance，the accuracy and recall rate of both algorithms are higher than 90%，and the performance of the KNN algorithm is better than that of the RF algorithm. Meanwhile，the two proposed methods greatly improve the accuracy rate over previous studies and provide a reliable way for excavation stability assessment.