عنوان مقاله [English]
In this paper, in order to numerically model the bond-slip behavior of embedded bars, an applicable procedure was proposed. To evaluate the efficiency of the proposed model, a credible experimental specimen was selected and modeled in Abaqus software. Comparing the numerical and experimental results of the specimen confirmed the acceptable accuracy of the proposed model. Afterwards, two monolithic and precast reinforced concrete beam-to-column connections were chosen from experimental tests and were numerically simulated. Investigation on the precast specimen showed that the required embedded length of longitudinal bars of beam was not considered. Hence, slippage of the longitudinal bars of the beam at the connection area led to degradation of connection strength. In order to consider this slippage in finite element modeling, the proposed approach was employed. Analytical results showed a suitable agreement with experimental ones and slippage of the beam bars was observed in the finite element analysis. Consequently, in order to prevent the slippage of beam bars, couplers at the end of the bars were used. Required area of the couplers was calculated as such to develop yielding in longitudinal beam bars. These couplers were added to the precast specimen and afterward, this specimen was named modified-precast specimen. Capacity of the modified-precast specimen including yielding load, ultimate load and ductility was improved in comparison with monolithic and precast specimen. However, the ratios of ultimate moment to flexural strength in modified-precast and monolithic specimens are approximately equal to one, which shows the formation of flexural plastic hinge in the beams of both modified-precast and monolithic specimens.