Confined Masonry Walls Retrofitted with FRP

Masonry buildings with confined walls have performed well during the past earthquakes. The same cannot be said for unreinforced masonry walls. In the former buildings, even when damage occurs, falling roofs and losses of life do not usually follow. This is because confined masonry walls have higher strength ductility and are more stable. The Iranian Seismic Code (Standard 2800) makes the use of horizontal and vertical ties mandatory for masonry buildings. In spite of this, such confined walls have not been studied sufficiently. In this study, the nonlinear behavior of masonry walls is examined using finite element discretization. From the two types of modeling that are commonly used for the study of masonry material, namely macro- and micro- modelings, the latter are employed here. This is because such a model can provide more detailed information. Micro- models are the best tools available for understanding the behavior of masonry structures. They can depict all the failure mechanisms of the system. The behavior of mortar joints and masonry unit-mortar interface is lumped into a set of discontinuous elements. In this way, each joint, consisting of mortar and two unit-mortar interfaces, is modeled by a zero-thickness interface element. In other words, the masonry structure is modeled by a set of elastic blocks bonded together with potential fracture/slip lines at the joints. The composite interface model includes a tension cut-off for mode I failure, a coulomb friction envelope for mode II failure and a cap mode for compressive failure. For modeling the behavior of concrete, a model suggested by Thorenfeldt and Hordijk is used. For the longitudinal reinforcing bars, the failure criterion is that of Von Mises. The hardening of steel is also considered. The interface between the reinforced concrete members and the masonry units panel is modeled by the coulomb friction model including a tension cut-off mode. A parametric study is conducted for confined masonry walls by changing with different dimensions, boundary conditions and loading patterns. The results indicate that failure has one of the two failure modes: diagonal tension or rocking. In cantilever walls with rather large heights compared to their length, the failure mode is rocking. In the other cases, diagonal tension failure mode occurs. The use of tie also affects the capacity of the wall. This can be considered in the design of masonry structures. The results of nonlinear analyses show that deboning does not occur between the ties and the body of masonry walls. Therefore, in analytical studies, the adjoining nodes for the two parts can be merged. Upon determination of failure modes, different patterns of FRP were investigated for the dominant mode in order to select the most suitable pattern. The FRP configuration patterns considered for strengthening the wall were: 1. FRP in vertical direction; 2. FRP at both ends in vertical direction; 3. FRP in diagonal direction; 4. FRP covering the whole surface of the wall. The last pattern was considered only for reference as it cannot be justified because of the increased cost. In confined masonry walls with diagonal tension or rocking mode, the best strengthening configuration is diagonal. The increase in the capacity of strengthened walls with rocking and diagonal tension failure modes depends directly on the amount of FRP. In fact, the capacity increases from 1.2 to 2 for the walls with rocking mode. For the walls with diagonal tension mode, this increase is from 1.5 to 3.