挫屈束制支撐鋼框補強既有鋼筋混凝土構架之研究

Abstract

隨著地震工程研究發展，以及規範的修改與演進，老舊鋼筋混凝土建築物常因不符合現行耐震規範之需求，須進行耐震補強以提升其耐震性能，而使用鋼斜撐框架補強能有效提升結構之耐震性能。目前鋼斜撐框架補強之工法中，對於鋼框架以及鋼筋混凝土構架兩者間之界面，大部份是以後置式錨栓連結，進行力量傳遞，由於後置式錨栓使用數量常較多，施工時之鑽孔需求亦增加。本研究使用挫屈束制支撐以及T型斷面之鋼框架進行補強，有別於常見之補強方式以後置式錨栓傳遞鋼框架之強度至鋼筋混凝土構架，而改以構架四個角落之水泥砂漿承壓塊做為力量傳遞之媒介，可大量減少後置式錨栓之使用量。本研究旨在探討此種補強方式下，構架之傳力機制及補強性能。本研究依此補強方式設計兩座鋼斜撐框架補強試體(BRB-S-WTF和BRB-D-WTF)及一座純鋼框架補強試體(WTF)進行試驗。試驗結果顯示，鋼斜撐框架補強可使構架側力強度持續成長至3%弧度側位移角，提升構架之側力強度達1060kN(約為原RC構架側力強度之3倍)，並持續變形至5%弧度側位移角時BRB核心才發生斷裂。補強構架於斜撐受壓時會對構架角隅附近造成壓力，此一壓力可能導致梁柱接頭破壞或在柱端產生似托架之破壞行為。本研究以軟化壓拉桿模型檢核接頭及托架行為之破壞，對於BRB-S-WTF試體，計算之梁柱接頭壓桿強度為810kN，BRB-S-WTF試體BRB之最大軸壓力為1025kN > 810kN，預測梁柱接頭會破壞；而BRB-D-WTF試體BRB之最大軸壓力為807kN < 810kN，預測梁柱接頭不會發生破壞，兩者皆符合試驗結果。另一方面，RC柱之混凝土之抗壓試驗強度為25MPa。若以軟化壓拉桿模型檢核柱之托架破壞，則BRB-S-WTF試體托架之壓桿強度需求為25.1MPa 25MPa，預測會發生托架破壞；而BRB-D-WTF試體托架之壓桿強度需求為19.5MPa < 25MPa，預測托架不會發生破壞，兩者皆符合試驗結果。故軟化壓拉桿模型能有效預測試體梁柱接頭及托架行為之破壞。本研究亦利用Abaqus有限元素模型分析結果所觀察到之傳力行為推導預測構架側力強度及各梁柱構件內力之簡化分析方法。由簡化分析方法計算BRB-S-WTF試體之側力強度為1120kN，實際試驗反應為1289kN；計算BRB-D-WTF試體計算之側力強度為973kN，實際試驗反應為1018kN，故簡化分析方法能有效預測構架補強後之側力強度，所得之構件內力亦可用於RC構件之強度檢核。本研究提出一種以挫屈束制支撐補強既有鋼筋混凝土構架之設計、檢核流程。

With the advancement in earthquake engineering technology and seismic provisions, many existing old reinforced concrete (RC) structures do not meet the current seismic standards. Seismic retrofit using steel braced frames has been found as a popular method to improve the seismic performance of RC buildings. For this purpose, closely-spaced post-installed anchors are widely adopted in the interface between the RC and steel frames to transfer the seismic loads. It often requires a substantial amount of on-site work of hole drilling and installation of anchors on the RC members. In this research, three identical strong-beam-weak-column RC frames are retrofitted, each with or without a diagonal buckling-restrained brace (BRB) incorporated into a steel braced frame using WT sections on four sides. Instead of applying the post-installed anchors to transfer the seismic loads, high-strength mortar bearing blocks at the four corners of the RC frame are constructed. The loads in the steel frame are transferred to the RC frame through the bearing blocks. This approach can reduce the usage of large amount of post-installed anchors. The purpose of this research is to evaluate the performance of the retrofitted frame, and investigate the load transfer mechanism between the RC frame and the steel frame. In this study, cyclic loading tests were applied on two RC frames retrofitted with steel braced frames using BRBs (BRB-S-WTF and BRB-D-WTF), and on one RC frame retrofitted with the steel frame without the brace. Test results show that, the lateral strength of BRB-D-WTF continued to increase until reaching 3% drift ratio, and developed 1060kN story shear (3 times of the bare RC frame) before the BRB core fractured at the 5% drift ratio. When the brace is in compression, it imposes a compressive force near the tip of the frame corner. This compressive force may cause an RC joint failure or a corbel-type failure in the RC column. In this research, these two failure modes are evaluated using the softened strut-and-tie model. The strength of the compressive strut in the beam-to-column joint is 810kN. In the BRB-S-WTF specimen, the estimated maximum brace compressive force is 1025kN. Thus, the beam-to-column joint failure was predicted. In the BRB-D-WTF specimen, the estimated maximum brace compressive force is 807kN. Thus, no beam-to-column joint failure was anticipated. The predictions match the test results. The concrete compressive strength of the RC column is 25MPa. In the BRB-S-WTF specimen, the strength demand on the corbel computed using the softened strut-and-tie model is 25.1MPa and approximately equals 25MPa. Thus, it predicted that the corbel failure might occur. In the BRB-D-WTF specimen, the strength demand on the corbel is 19.5MPa and less than 25MPa. The model predicted that the corbel failure should not occur. These two predictions also match the test results. Therefore, it appears that the softened strut-and-tie model is effective in predicting the joint failure and corbel failure of the RC frame. In this research, a simplified method for estimating the lateral strength of the retrofitted structure and the internal forces of RC beams and columns is developed from observing the load transfer mechanism found in the Abaqus finite element model (FEM) analysis results. The estimated lateral strengths for the BRB-S-WTF and BRB-D-WTF specimens are 1120kN and 973kN, predicted the experimental results of 1289kN and 1018kN, respectively. This confirms that the proposed simplified method is effective for estimation of the lateral strength of the retrofitted frame, and computing the estimated internal force demands on RC frame members. This study proposes a design procedure for retrofitting existing RC frames with buckling-restrained braced frames.