Summary: | 博士 === 國立臺灣科技大學 === 營建工程系 === 93 === According to the post-earthquake reconnaissance of the Chi-Chi
earthquake, the low-rise RC buildings were highly vulnerable to the earthquake.
On the other hand, a substantial amount of the low-rise RC buildings with
sufficient partition walls survived from the Chi-Chi earthquake. However, the
present codes do not provide sufficient design information related to the
prediction of behavior of shear walls, especially in the estimation of deflection.
For example, it is well known that the deflection of low-rise frames infilled
with walls is dominated by shear deflection, while as only the deflection
induced by bending moment is clearly stated in codes. Therefore, the
objective of this research is to develop an analytical model for the estimation of
shear deflection and to establish a method for the prediction of the
load-deflection curves of the low-rise frame with partition walls.
In the proposed model, the strain field of the wall web is estimated
according to the softened strut-and-tie model. Since the constraints due to
force equilibrium, constitutive law and strain compatibility are satisfied, thus a
unique analytical solution can be obtained. Compared with test data in
literature, it is found that the deflection at ultimate strength could be reasonably
predicted by the proposed model and that the estimated shear deflection is
about 80-95% of the total deflection predicted.
A bi-linear load-deflection curve, consisted of cracking and ultimate point,
is recommended to simulate the lateral response of RC walls. As compared
with the shear wall database, the proposed model predicts reasonably well. A tri-linear load-deflection curve whose control points were the cracking,
yielding and ultimate stages is also suggested for the estimation of column
behavior. Compared with test data in literature, the prediction is also
reasonable.
For the further verification of the proposed model, the tests performed in
the National Center For Research On Earthquake Engineering were compared.
The NCREE data bank consists of 23 frames in four categories, i.e. pure frame,
frame with partition walls, frame with partition walls using the Carbon Fiber
Reinforced Polymer (CFRP) and the two-story and two-bay non-ductile frame
with RC walls throughout the second floor. As shown in the study, the
aforementioned test results can be reasonably predicted by the models
suggested.
|