振动台实验和动态响应地震后的钢筋混凝土建筑英文文献和中文翻译(2)
SCT85-10 excitation (corresponding to PGA 0.194g, the input in calculation). The test model had higher acceleration responses than the prototype building. The difference between the calculated and the measured results can be attributed to the similarity between the reduced-scale model and the prototype building in nonlinear status, and to the gap between the analysis model and the real building. In the nonlinear stage, the model and the prototype building had different stiffness degradation and vibration properties. The floor slabs with increased thickness in the test building model had a greater effect on its stiffness than those in the prototype building. In analysis, the effect of the slab and slab bars to beam members were not included, and assuming a rigid floor, did not reflect the effect of floor slab deformation. Again, the reinforced concrete column member bad evently distributed reinforcing bars around the four sides, so it did not have a clear yielding point. The method used to determine the yielding capacity and post-yielding stiffness of the members influenced the member behavior, especially at the yielding and post- yielding stages. In the analysis model, the nonlinear shear deformation of beams, columns and beam- column joints could not be taken into account, while the prototype building had diagonal cracks on these members. Additionally, the analysis model was not able to consider concrete crush and spall off. These factors may have reduced the calculated results when compared with the actual seismic response of the prototype building. 5 Conclusions Shaking table test and nonlinear response analysis are two important approaches used to understand the seismic response and damage mechanism of structures subjected to strong earthquakes. In universities and research institutes in China, shaking table facilities have been established to perform experimental research. Through evaluating the shaking table test results presented in this paper and from some other research reports, the following conclusions are obtained. The shaking table test results are affected by the specimen size, similitude design, input and excitation control, and precision of the data acquisition system. The test of a carefully designed and constructed one- tenth-scale RC building specimen has shown that the damage pattern of RC buildings observed in past earthquakes can be reproduced to a certain degree of reliability in the laboratory. The studies also indicate the possibility of using a sophisticated
3-D nonlinear numerical analysis to simulate the dynamic response of RC frame buildings for minor and moderate earthquakes, but matching the response to a severe earthquake is difficult.