| Summary: | 碩士 === 國立中正大學 === 地震研究所 === 107 === In order to avoid the casualties caused by damaged buildings during strong earthquakes, it's important and essential to understand the seismic design specifications of buildings. The motion of a building depends on the interferometry of building and ground motion, the coupling between building and ground, and the mechanical properties of the building. We applied deconvolution to the motion recorded to separate the building response. Deconvolution interferometry is a powerful and convenient tool which allow us to extract structural parameters (e.g., travel wave velocity, resonant frequency and attenuation parameter) from seismic records. By applying deconvolution to earthquake records, it can provide a complete response of the building and understanding the motion pattern of the structure under strong motion. On the other hand, ambient noise can obtain continuous result with stable sources. Therefore, we can check medium's state by tracking travel wave velocity, resonant frequency and the attenuation parameter (Q-value) to avoid damage of building in the next mega-earthquake.
We deployed a seismograph array in the library of National Chung Cheng University, Taiwan, which recorded earthquakes and few months of ambient noise record. The travel wave velocity from the seismic record was calculated to be 225.80±22.97 m/s. We also used seismic record to calculate the attenuation parameter and it shown with high uncertainty Q=18.33±10.01, we believed was refer to the instrument direction. On the other hand, we estimated the similar travel wave velocity 292.18±6.70 m/s from ambient noise record for the entire building as well. In advance, we divided the structure into lower and higher floors to calculate the travel wave velocity from the upward and downward propagation. We found that the travel wave velocities of the lower and higher floors are significantly different but the former exhibit stable variation. For the monitoring phase, calculating the velocity variation from ambient noise depends on stacking signal to gain better S/N ratio and more convergent result. Therefore, for ambient noise calculation, we tried different stacking time-lengths and calculated the related travel wave velocities which are similar to the analysis in earthquakes. From the above results, we concluded that the physical parameters of the building are stable and implied that the structure is in a healthy condition. Finally, we hope these results would be helpful to build a long-term monitoring of building's healthy status and the assessment of seismic hazard.
|
|---|
