تجزیه و تحلیل حساسیت از تیرهای بتن مسلح تقویت شده با لامینیت های FRP

Numerical procedures are proposed to predict the failure of reinforced concrete (RC) beams strengthened in flexure with fiber-reinforced polymeric (FRP) laminates. The framework of damage mechanics was used during the modeling. Numerical results were validated against experimental data obtained from 19 beams strengthened with different types of FRP. These beams failed by concrete crushing, cover failure and plate debonding. The numerical models were capable of predicting the experimentally observed load–deflection, failure load and failure modes. The sensitivity of the numerical results was studied. In particular, the effect of the concrete constitutive behavior and different modeling considerations was evaluated. It was found that the fracture energy of the concrete–repair interface plays a central part in predicting plate-debonding failures.

Strengthening reinforced concrete (RC) beams with fiber-reinforced polymer (FRP) composites is becoming an attractive alternative for the construction industry. These laminates offer all the advantages of composite materials, such as a low volume to weight ratio, and a high strength-to-weight ratio. Experimental tests of FRP-strengthened beams have identified a number of possible failure modes: tensile rupture of the composite laminate, debonding failure between the FRP laminate and the concrete substrate, concrete cover failure, and concrete crushing [1]. Although experimental data is valuable in understanding the behavior of this strengthening system, analytical and numerical solutions are needed to further comprehend and predict the behavior and failure mechanism of the strengthened beams. Numerical modeling of FRP-strengthened beams represents a formidable challenge because other aspects—such as loading sequence, construction procedure, nonlinear material behavior, crack propagation and residual stresses—may have a significant impact on the results obtained in such an approach. Different modeling approaches, partially incorporating these aspects, have been proposed up to now; but sensitivity analyses are needed in order to identify the principal parameters controlling the numerical response of FRP-strengthened beams and to simplify the modeling procedures currently available. Among different modeling approaches proposed to date, Ziraba and Baluch [2] presented a nonlinear finite-element code to simulate the global behavior, up to failure, of RC beams externally reinforced. They were able to predict the behavior of strengthened members subjected to arbitrary load histories previous to strengthening. In this approach, the interfacial behavior between concrete and the internal and the external reinforcement was modeled using six-node interface elements; nine-node Lagrangian elements were used for modeling the concrete and external laminate, whereas a three-node element represented the internal reinforcement. Arduini et al. [3] conducted a finite-element analysis of eight FRP-strengthened beams using the commercial code ABAQUS. During this study, the concrete was modeled using the smeared crack approach, and a perfect bond was assumed between the FRP and the concrete. This analysis was limited to beams with a monotonic loading history and without damage prior to FRP installation. Wong and Vecchio [4] used link and contact elements following linear and elastoplastic bond laws in order to model the bond-slip behavior at the bond interface of FRP-strengthened beams. A two-dimensional nonlinear finite-element code based on the Modified Compression Field Theory was used during the analyses. The procedure was able to predict the load–deflection behavior of FRP-strengthened beams failing by laminate debonding only when using an elastoplastic bond-slip behavior. However, the authors stated that a more clearly defined constitutive relationship for the bond elements must be developed to further improve and validate the model capability. From these previous studies, it can be concluded that due to the challenges faced during the numerical analysis of FRP-strengthened beams, most of the research in this area has been devoted to improving the modeling techniques. From a practical point of view, it is also necessary to identify the principal material properties affecting the outcome of the numerical simulations; evaluate the sensitivity of the numerical results to variations in these material properties; and to recommend procedures for estimating such properties when experimental data are not available. This study aims at fulfilling these needs. For this purpose, numerical results were validated against experimental data obtained from 19 beams strengthened with different types of FRP, and their sensitivity to concrete constitutive behavior and different modeling considerations was evaluated. In addition, practical recommendations are given for the modeling of FRP, epoxy, selecting model parameters and element size.

The behavior of reinforced concrete beams strengthened with FRP laminates was simulated using the finite-element method (FEM). The sensitivity of the numerical results was studied. In particular, the effect of the concrete constitutive behavior and different modeling considerations was evaluated. Practical recommendations are given for the modeling of FRP, epoxy, selecting model parameters and element size. Conclusions derived from this study are as follows: • The FEM models were capable of predicting the load–deflection behavior of both reinforced concrete (RC) beams and FRP-strengthened RC beams. The failure modes predicted (concrete crushing, plate debonding and cover failure) were the same as those obtained experimentally. • The numerical simulations indicate that the concrete tensile strength does not constitute the unique failure criterion for predicting plate-debonding failure of strengthened RC beams. • The fracture energy of the concrete–epoxy interface is needed to accurately predict plate-debonding failures. This finding agrees with experimental observations, indicating that this phenomenon is due to a crack propagating along the concrete–epoxy interface. • Peak loads and deflections are slightly affected by the type of concrete softening used. From a practical point of view, these effects are negligible. • No significant changes were observed in the global response of the CFRP-strengthened beams when different modeling approaches were used for the FRP laminate. • The numerical results were insensitive to the mesh geometry, which demonstrates the objectivity of the modeling approach used during this study. • The modeling of the epoxy has a minor effect in the overall response of strengthened RC beams, even for beams failing by plate debonding. Further experimental and analytical work is currently underway to obtain the fracture energy of the concrete–epoxy interface.