Strategies for Selecting Infilled Frame Models for Seismic Performance: Meso and Macro Models

When laboratory testing is inconvenient, numerical modelling of infilled frames is required in order to represent the elastic and inelastic behaviour of these structures. However, the choice of model is very complex, as an infilled frame is made up of various materials such as bricks, mortar, concrete and reinforcements. The most accurate approach to modelling is a technique that is capable of representing the actual behaviour of the structure. In this study, we carry out numerical analysis and compare two modelling techniques, meso and macro modelling, in terms of the structural response produced. A meso model of an infilled frame has the advantage of capturing the detailed local behaviour of the panels, but has disadvantages such as complexity and high computational load. The accuracy of the structural response modelled with a meso approach is shown to have a similarity of 98% with the test results. In contrast, macro modelling using diagonal struts can significantly reduce the analysis time without compromising the accuracy of the results. This method can also accurately reproduce the nonlinear global structural behaviour. The disadvantage of this type of model is that the resulting structural response cannot represent the local damage to the masonry panel in detail. The similarity between the behaviour of the macro model and the experimental results is shown to be 104%. We find that a meso model of the infilled frame is more responsive to the initial variables derived from test results for the materials used in the structure. This is obvious in the initial stages of creating a meso model based on test data on masonry panels, as the structural response is much closer to the test results than that of the macro model. This may be because the geometric and mechanical behaviour of the meso model is closer to the actual conditions. The damaged states of the infilled frame based on the local behaviour in the meso model are related to the structural-nonstructural element performance criteria used in macro modelling.