使用非線性黏性阻尼器之彈性結構受近斷層地震作用之分析與設計研究

Abstract

對於結構物減震設計而言，結構物整體阻尼比大小和結構物之地震反應息息相關，結構物傳統阻尼比的概念主要來自線性系統，因此對於含非線性阻尼器之構架理論上無法由傳統之阻尼比( )定義，目前常用於計算含非線性黏性阻尼器構架之等效阻尼比方法有二：等能量法及等功率法。從本研究對結構物裝設非線性黏性阻尼器並受近斷層地震歷時作用下的分析結果整理，可以得知，於結構物裝設由等能量法求得的阻尼常數之非線性阻尼器，可以使結構物之反應有較好的等效線性效果。 對於裝設線性阻尼器之結構物，ASCE7-10規範對於結構物有效阻尼比異於5%時提出適當的阻尼修正係數(B值)，配合5%彈性加速度反應譜求得高阻尼比之彈性反應譜，但是ASCE7-10規範不建議該規範之B值使用於位於軟弱土層及含近斷層地震特性作用之結構物分析，因此對於此類型之結構物分別提出譜加速度阻尼修正係數( )及譜位移阻尼修正係數( )，藉以改善使用ASCE7-10規範B值評估單自由度結構物反應於某些條件下過於保守或過於不保守的情況。 ASCE7-10規範中利用反應譜分析(RSA)及等值側力分析(ELF)方法評估結構物裝設阻尼器受地震力作用下的各種反應最大值，以SAP2000非線性模態疊加之非線性動力分析結果作為標準進行比較。為比較現有規範B值及 、 在評估結構物受近斷層地震下反應最大值的準確度，因此使用5層樓、15層樓及34層樓構架進行各種反應之分析討論，除了使用規範現有的兩種評估方式外，也對現有規範的評估公式作些許的修正，但結果發現，不論上述何種評估方式的使用，對於結構物樓層最大絕對加速度反應在低樓層部分皆有低估的現象，在高樓結構及結構物高阻尼比的情況下尤其明顯。 為解決樓層最大絕對加速度反應在低樓層低估的現象，本研究提出以單自由度歷時分析為基礎的多自由度系統簡化分析法，希冀可以準確的評估結構物樓層最大加速度反應。

ASCE7-10 provides a reduction factor B for engineers to determine the seismic demands of buildings equipped with energy dissipation systems, such as viscous dampers. However, ASCE7-10 does not suggest the use of the factor B on buildings located at soft soil sites and those subjected to ground motions with near-fault characteristics. For buildings subjected to near-fault ground shaking, reduction factors for spectral acceleration ( ) and displacement ( ) are developed in this study to improve the accuracy of the design procedures of ASCE 7-10. In this study, a series of nonlinear response-history analyses (RHA) are conducted for sample 5-, 15- and 34-story buildings equipped with nonlinear viscous dampers. The results of RHA are used to benchmark those computed using the two design procedures provided in ASCE 7-10, namely, Response Spectrum Procedure (RSA) and Equivalent Lateral Force Procedure (ELF). The pros and cons of RSA, ELF, and the B values provided in ASCE 7-10 are discussed. A new method is proposed in this study using multiple single degree-of-freedom linear RHA to capture the responses of buildings with nonlinear viscous dampers and subjected to near-fault ground motions. The new method provides more accurate results than the RSA and ELF of ASCE 7-10, especially in the prediction of peak floor acceleration.