چكيده به لاتين
Concrete under uniaxial compressive loading, at low stresses (up to about 50% compressive strength of concrete), has a lateral strain of about 15 to 20% of the axial strain. However, by increasing the amount of uniaxial stress, micro-cracks grow and develop. Due to the presence of these micro-cracks, the lateral strain increases. As the stress reaches the final resistance range and beyond, the Poisson's ratio increases rapidly and rupture occurs.
Columns are one of the most sensitive members of any structure and their predominant load is generally compressive. In the lateral confined concrete columns, under compressive loading, internal cracks are formed in the concrete and occur lateral expansion. In this case, the wrapped jacket around the column, applies lateral stresses to the concrete core. The amount of these stresses, increase with more expansion of concrete core. In other words, the amount of lateral stresses is a function of the lateral expansion of the concrete core. On the other hand, the lateral expansion of concrete depends on the value of the Poisson's ratio. Therefore, in this study, in order to investigate the effect of Poisson's ratio variations, reinforced and unreinforced concrete columns confined by FRP with square-rectangular cross section, have been investigated using finite element method.
This research presents a realistic as well as detailed 3D finite element model within the framework of concrete damage plasticity model in ABAQUS software to predict the behavior of CFRP-confined concrete short columns with various rectangular and square cross-sectional areas subjected to compressive monotonic loading. The advantage of the proposed method is that unlike other similar works, instead of stress–strain curve of confined concrete, the unconfined one is used here as the uniaxial compressive behavior, and the real effects of FRP confinement, by considering Poisson's ratio variations and modeling techniques are considered. In fact, the actual distribution of confinement pressure, by correctly simulating the lateral expansion of concrete (which is related to the variations in the Poisson's ratio of concrete) and considering the FRP-hoop strain are included in the numerical model. It is noteworthy that two methods have been proposed to consider variations in the Poisson's ratio of concrete. The first method, which is a relatively approximate method, because of ABAQUS finite element software is not capable directly to considering Poisson's ratio variations, and because of their simplicity and lack of need for scripting, is presented. In addition, the second method, which is an exact method, has been implemented using USDFLD subroutine in ABAQUS software.
Comparing the results of numerical analyzes with the collected experimental results from reliable sources, show that in the approximate method, the average of errors (AAE) and the average of standard deviation of error (SD), are equal to 14% and 6.16%, respectively. Also in this method, the average of the numerical to the experimental strain ratio is equal to 1.04. On the other hand, in the exact method, the average of errors (AAE) and the average of standard deviation of error (SD) are equal to 12.33% and 4.97%, respectively. Also in this method, the average of the numerical to the experimental strain ratio is equal to 1.00. Therefore, it can be said that both proposed methods have acceptable accuracy in predicting the axial stress-strain behavior of FRP-confined concrete rectangular columns.
