多種隔震位移系統位移容量設計法之研究

Abstract

台灣位於歐亞板塊與菲律賓海板塊交界處，地質構造複雜，地震活動頻繁，歷經如1999年集集地震、2016年美濃地震及近年的花蓮地震等多次重大災害事件，造成嚴重人員傷亡與經濟損失。隨著都市化進程加速，使得建築物應具備減少震後功能中斷的韌性設計。為因應此挑戰，結構控制技術逐漸成為提升結構耐震性能的重要手段，特別是在高重要性建物與基礎設施領域中應用日益廣泛。 結構控制依其作動方式可分為主動、半主動與被動控制系統，其中被動控制系統因具備不需外部能源、機構簡單與穩定性高等優點，最為實務界所接受。隔震系統即屬於一種被動控制裝置，其透過在建築物基礎與上部結構之間設置柔性隔震層，有效延長結構的自然週期，使其遠離地震能量主要集中的頻帶，進而顯著降低地震反應，提升上部結構安全性與耐震韌性。另一方面，慣性質量系統（Inerter System）為近年新興之控制元件，其主要藉由兩端相對加速度產生反作用力，模擬出虛擬質量效應，進而改變結構的動力特性與頻率響應，在不需增加實體質量的情況下，有效抑制結構位移或加速度，於遠域地震作用下特別展現出良好的控制性能。近年來，將慣質裝置整合於傳統隔震系統之中，發展出具備複合控制機制之線性慣質隔震系統，成為提升現行隔震技術之新興方向。 本研究針對不同類型之隔震系統進行位移容量設計法之探討，提出一套系統化、實務可行且考慮地震輸入特性之設計流程。研究動機源於現行隔震系統設計流程中常見之非線性調整與重複遞迴計算；而地震輸入中近斷層脈衝效應常被忽略，進一步引發設計誤差。為提升設計效率與準確性，本研究採用蒙地卡羅法隨機生成多組上部結構參數，結合主動設計概念推導對應之隔震層力學參數，並建立三種隔震系統模型：線性隔震系統、帶有線性慣質系統之隔震系統、以及具雙線性遲滯行為之鉛心橡膠支承系統（LRB）用以比較隔震系統實際考慮支承的非線性行為以及加裝慣性質量元件後系統反應的變化。 為避免原始地震特性對分析結果造成偏誤，遠域地震輸入透過白噪音結合Kanai-Tajimi與Clough-Penzien地震模型模擬生成，近斷層地震則以數學模型擬合具代表性速度脈衝之輸入歷時。分析以MATLAB進行歷時反應模擬，觀察各類隔震系統於不同地震情境下之最大位移反應，並以第95個百分位最大位移作為位移容量評估依據。 模擬結果顯示，線性隔震系統在遠域地震下呈現線性增長與平台段之趨勢，加入慣質裝置可進一步降低位移需求；但於近斷層地震下，平台段消失且最大位移呈持續上升，慣質系統甚至可能導致不利影響。LRB系統具明顯非線性遲滯消能效果，其位移容量主要受設計變形能力控制。最後，本研究分別針對遠域與近斷層地震情境提出簡化位移容量設計公式，並考量質量比與有效阻尼比等參數影響，以期提供工程實務上高效、可靠之隔震設計依據。

Taiwan is located at the boundary between the Eurasian Plate and the Philippine Sea Plate, characterized by complex geological structures and frequent seismic activity. Major seismic events such as the 1999 Chi-Chi earthquake, the 2016 Meinong earthquake, and the recent Hualien earthquake have caused significant casualties and economic losses. As urbanization accelerates, the demand for resilient design that minimizes post-earthquake functional disruption has become increasingly critical. To address these challenges, structural control technologies have emerged as important tools for enhancing seismic performance, with growing applications in critical buildings and infrastructure. Structural control systems are generally classified as active, semi-active, or passive. Among them, passive control systems are most widely adopted in practice due to their simplicity, reliability, and independence from external power sources. Base isolation is a common form of passive control that introduces a flexible isolation layer between the superstructure and the foundation, effectively lengthening the natural period of the structure. This shifts the dynamic response away from the dominant energy range of seismic excitation, significantly reducing structural response and enhancing seismic resilience. In recent years, inerter systems have emerged as novel control devices that generate reactive forces proportional to relative acceleration across their terminals, simulating virtual mass effects. These systems can alter the dynamic properties and frequency response of structures without significantly increasing physical mass and have demonstrated promising control effectiveness under far-field ground motions. The integration of inerter devices into traditional base isolation systems has led to the development of linear isolation systems with inerters, representing a new direction in seismic protection technology. This study investigates displacement capacity design methods for various types of isolation systems and proposes a systematic and practical design framework that accounts for different ground motion characteristics. The motivation stems from the nonlinear iterations commonly encountered in current design procedures and the frequent neglect of near-fault pulse effects, which may lead to underestimation of seismic demand. To improve design efficiency and accuracy, this research employs the Monte Carlo method to randomly generate multiple superstructure configurations and applies an active control-based design concept to derive the mechanical parameters of the isolation layer. Three isolation models are constructed for comparison: linear base isolation systems, linear isolation systems with inerters, and lead rubber bearing (LRB) systems with bilinear hysteresis behavior. These models enable evaluation of the effects of nonlinearity and inertial mass integration on system responses. To eliminate bias from specific recorded ground motions, far-field inputs are simulated using white noise filtered through Kanai-Tajimi and Clough-Penzien filters. Near-fault inputs are generated using mathematical functions that replicate representative velocity pulses. Time-history analyses are performed in MATLAB to evaluate the maximum displacement responses of each isolation system under different seismic scenarios. The 95th percentile of the maximum displacement is used as the reference for displacement capacity. Simulation results indicate that linear isolation systems under far-field ground motions exhibit a trend of initial linear growth followed by a plateau, and the addition of inertial devices further reduces displacement demand. However, under near-fault ground motions, the plateau region vanishes and displacement increases continuously, with inertial systems potentially leading to adverse effects. The LRB system, exhibiting strong nonlinear hysteretic energy dissipation, shows displacement capacity primarily governed by its deformation limits. Finally, this study proposes simplified displacement capacity design formulas for both far-field and near-fault scenarios, incorporating the influence of mass ratio and effective damping ratio, with the aim of providing a practical, efficient, and reliable reference for engineering design of base-isolated structures.