DEM Algorithm for Progressive Collapse Simulation of Single-Layer Reticulated Domes under Multi-Support Excitation

In this paper, the Member Discrete Element Method (MDEM) is modified and perfected for three aspects: the algorithm itself, loading and computational efficiency, and to accurately and quantitatively simulate the progressive collapse for large-span spatial steel structures. In addition, the corresponding computational programs are compiled. First, from the perspective of the method, a meshing principle for discrete element models is determined, a treatment for material nonlinearity and strain rate effect is proposed, and a damping model is established. Next, the Displacement Method is introduced to determine the multi-support excitation for the MDEM, and then motion equations of particles under multi-support excitation are derived. On this basis, the specific process of gravitational field loading is presented. Furthermore, parallel implementation strategies for the MDEM based on OpenMP are constructed. Finally, the collapse simulation of a 1/3.5-scaled single-layer reticulated dome shaking table test model under multi-support excitation is carried out. The comparison demonstrates that the ultimate load and failure mode as well as the complete collapse time of the numerical results are consistent with the experimentally measured responses, and the configuration variations from member buckling and local depression until collapse failure are fully captured. Moreover, the displacement time-history curves obtained using MDEM are almost identical to the experimental measurements, and there is a nuance only in the amplitude. It is verified that MDEM is capable of precisely addressing the collapse failure for large-span spatial steel structures. Additionally, the failure mechanism for structures of this type is naturally revealed.     

The Coupled Influence of Relative Density,CSR, Plasticity and Content of Fines on Cyclic Liquefaction Resistance of Sands

In this study, stress-controlled cyclic simple shear tests were performed on sand specimens with up to 10% silt or clay contents, and coupled effects of plasticity, fines content (FC), relative density (D_R) and CSR (cyclic stress ratio) were investigated. The results demonstrated that for sands with low fines content, reasonable trends were obtained when the packing index control parameter was selected as D_R. Also, clean sand specimens demonstrated highest liquefaction strength compared with that of sands with fines up to 10% FC. The effect of fines’ plasticity became apparent as the FC increases and CSR reduces in relatively denser specimens.     

The Shaking Frequency Effect in the Dynamics of a Model Sandy Layer

Experimental results showing the frequency-dependent behavior of both dry and saturated sandy layers subjected to a horizontal excitation on a shaking table are presented. The largest settlements of a dry layer correspond to two specific frequencies. In the case of a saturated layer, there is a single peak frequency corresponding to the largest depth of sinking of a measuring plate in liquefied subsoil. The first peak of settlements coincides with the single peak of sinking in liquefied soil. The eigenfrequencies of the layer were estimated. A modification of the compaction law was proposed for low shaking frequencies.     

Real-Time Dynamic Hybrid Testing Coupling Finite Element and Shaking Table

In this article, a Simulink simulation block with the finite element function is developed on the basis of S-function and implemented as the numerical substructure of real-time dynamic hybrid testing. Thereby, a real-time dynamic hybrid testing system coupling finite element calculation and shaking table testing is achieved. Using the developed system, a shear frame mounted on the soil foundation is tested, in which the shear frame is simulated as the physical model and the foundation is simulated as the finite element model with 132 degrees of freedom. Several cases of the dynamic behavior of soil-structure interaction are studied.     

Experimental and Analytical Study of Seismic Soil-Pile-Structure Interaction in Layered Soil Half-Space

The dynamic interaction of pile foundations, embedded in a horizontally stratified soil profile, with superstructures under low to moderate earthquake excitation can be handled in different ways. In this article, the soil-pile-superstructure dynamic interaction problem has been investigated using the coupled finite element-boundary element method. Comparison with shaking table experiments of a small scale model pile shows a good correlation with the proposed method in terms of the kinematic response of pile foundations and the structural response. A parametric study of the proposed model has yielded important results essentially concerning the amplification factors of the pile foundation and the superstructure.     

Impact Stiffness of the Contact-Element Models for the Pounding Analysis of Highway Bridges: Experimental Evaluation

For the requirement of pounding analysis of highway bridges, how to properly choose the impact stiffness has become a primary issue for an achieving accurate result. This article presents an evaluation test of the impact stiffness of four types of contact-element models based on the shaking table test results of a steel highway bridge model. The analytical results indicate that the theoretical impact parameters are significantly larger than the identified values because the assumptions for deriving those models cannot match the actual impact conditions. The possible reasons causing those differences are discussed at the end of this study.     

Study on a Simplified Calculation Method for Hydrodynamic Pressure to Slender Structures Under Earthquakes

Based on Morison hydrodynamic force theory, a simplified calculation method for hydrodynamic force to slender structures is proposed. A typical cylindrical deep water pile was chosen for study and the simplified method was used to analyze the influence of hydrodynamic pressure on the pile. Then the finite element method was used to study the dynamic characteristics of the model, and the results from both methods were compared. A comparison was also made between the shaking table test on the south tower of the Nanjing Yangtze 3rd River Bridge and the results from the simplified method.     

Full-Scale Experimental Verification of Soft-Story-Only Retrofits of Wood-Frame Buildings using Hybrid Testing

The FEMA P-807 Guidelines were developed for retrofitting soft-story wood-frame buildings based on existing data, and the method had not been verified through full-scale experimental testing. This article presents two different retrofit designs based directly on the FEMA P-807 Guidelines that were examined at several different seismic intensity levels. The effects of the retrofits on damage to the upper stories were investigated. The results from the hybrid testing verify that designs following the FEMA P-807 Guidelines meet specified performance levels and appear to successfully prevent collapse at significantly higher seismic intensity levels well beyond for which they were designed. Based on the test results presented in this article, it is recommended that the soft-story-only retrofit procedure can be followed when financial or other constraints limit the retrofit from bringing the soft-story building up to current code or applying performance-based procedures.     

Shake Table Test and Numerical Analysis of a Bridge Model Supported on Elastomeric Pad Bearings

Elastomeric pad bearings are widely applied in short- to medium-span girder bridges in China, with the superstructure restrained by reinforced concrete (RC) shear keys in the transverse direction. Field investigations after the 2008 Wenchuan earthquake reveal that bearing systems had suffered the most serious damage, such as span falling, bearing displaced, and shear key failure, while the piers and foundations underwent minor damage. As part of a major study on damage mechanism and displacement control method for short- to medium-span bridges suffered in Wenchuan earthquake, a 1:4 scale, two-span bridge model supported on elastomeric pad bearings were recently tested on shake tables at Tongji University, Shanghai. The bridge model was subjected to increasing levels of four seismic excitations possessing different spectral characteristics. Two restraint systems with and without the restraint of RC shear keys were tested. A comprehensive analytical modeling of the test systems was also performed using OpenSees. The experimental results confirmed that for the typical bridges on elastomeric pad bearings without RC shear keys, the sliding effect of the elastomeric pad bearings plays an important role in isolation of ground motions and, however, lead to lager bearing displacement that consequently increases the seismic risk of fall of span, especially under earthquakes that contain significant mid-period contents or velocity pulse components. It is suggested from the test results that RC shear keys should be elaborately designed in order to achieve a balance between isolation efficiency and bearing displacement. Good correlation between the analytical and the experimental data indicates that the analytical models for the bearing and RC shear key as well as other modeling assumptions were appropriate.     

Comparison of Numerical and Experimental Seismic Responses of FREI-Supported Un-reinforced Brick Masonry Model Building

Comparative study of numerically and experimentally obtained seismic responses of un-reinforced masonry building supported on in-house designed un-bonded fibre reinforced elastomeric isolator (U-FREI) are presented in this article. The effectiveness of U-FREI is established very clearly in terms of controlled dynamic response of the model building. Experimental studies are carried out on a shake table with elaborate instrumentations for measurement of acceleration and displacements at different floor levels. Numerical study of the model building supported on U-FREI is carried out to compare the results with experimental investigation. Multi-linear pivot hysteretic plasticity model is used to simulate the behavior of FREI, while plate elements are used for brick-masonry walls. Experimentally obtained force-displacement curves of FREI are used for defining the properties of multi-linear model representing FREI. The dynamic responses obtained from the numerical studies are compared with those from experimental investigations. This study indicates that the seismic responses of building supported on U-FREI can be numerically evaluated with quite reasonable accuracy. A good numerical model can be judiciously used at the preliminary design stage, followed by actual testing and construction of the base isolated building.