Taiwan is located in a highly seismic zone and the historical “Dieh-Dou” timber buildings, constructed without following any code or standard, are prone to collapse under earthquake. These buildings are unique and represent the culture, heritage and art of Taiwan, therefore need to be preserved while minimizing unnecessary intervention that could damage their authenticity. <br /> This research comprises a thorough investigation on the parameters influencing the seismic vulnerability of the Dieh-Dou timber frames in Taiwan, and propose a methodology of assessment and a strategy for strengthening validated through experimental testing and numerical analysis. <br /> After review existing literature and post-earthquake surveys, the failure modes of the buildings are identified, showing that the dislocation of the elements of the frame from the joints is the primary source of damage. An experimental investigation is carried out comprising both rotational and translational tests on full scale joint specimens which, together with a parametric study undertaken with an appropriate FE simulation, demonstrates how both the rotational and translational stiffness of the joints play a key role in defining the behaviour of these structures. <br /> Lateral force, response spectrum, and step-by-step pushover analyses are performed and compared with the post-earthquake survey of two Dieh-Dou buildings seriously affected by the 1999 Chi-Chi earthquake. The results show that the proposed FE modelling can successfully be employed to assess the vulnerability of the frames. <br /> Based on a damage level approach, an assessment methodology is suggested that would allow to optimisation of the strengthening strategy, permitting protection these precious structures from future earthquakes while avoiding unnecessary interventions.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:503662 |
Date | January 2009 |
Creators | Tsai, Pin-Hui |
Contributors | D'Ayala, Dina ; Walker, Peter |
Publisher | University of Bath |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Page generated in 0.0955 seconds