Numerical models for the robustness assessment of reinforced concrete framed buildings

Cast in situ reinforced concrete frame is one of the most common options for civil buildings. Although earliest common usages of this solution date back to second half of 19th century, research activity is constantly developing to investigate several aspects, especially about nonlinear behaviour of reinforced concrete structures. Structural robustness of buildings is actually one of the key issues faced by the international scientific community. This expression is used to indicate a desirable property of a structure that allows it to withstand an accidental event, preventing progressive and/or disproportionate collapse. Interest in this topic has been growing rapidly after the collapse of Ronan Point Apartment Tower in Newham, East London, in 1968, when a gas explosion destroyed a loadbearing concrete panel causing the collapse of an entire corner of the building. Although the issue of progressive collapse of multi-storey frames has been widely studied in the last decades, according to the literature review, the actual structural response after a localised failure has not yet been fully understood. Besides, many design guidelines for preventing progressive collapse denote a lack of adequate theoretical supports. Several technics have been developed to evaluate the response after an accidental situation. In Europe, EN1991-1-7 has introduced the notional removal design strategy. This approach establishes that a building should be checked to ensure that upon the notional removal of each column or each beam supporting a column, or any nominal section of load-bearing wall, one at a time in each storey of the building, the structure remains stable exhibiting only localised failure. Currently one of the main solution to ensure robustness consists in tying together structural members by using continuous reinforcement. In this context, the designer is required to evaluate the global structural response, then the role played by the floor-system becomes crucial. However, this operation involves longer times for modelling and analyses. The study here presented is articulated through several points. The initial intent is to develop simplified models of the floor-system able to simulate its behaviour, to obtain accurate results through a more efficient modelling. Different numerical models will be presented. These will focus on distinct simplification levels, depending on the finite elements adopted. The codes used for nonlinear numerical analyses have been previously tested and validated on experimental tests on 2D and 3D specimens and both static and dynamic analyses. The second aim is to evaluate the effectiveness of different floor-system typologies on the global behaviour of reinforced concrete frames. Two typical reinforced concrete solutions are tested: the first exploits a bidirectional slab, while the second uses monodirectional joists with a collaborating slab. To compare the results, a structure with features common to most of the reinforced concrete buildings has been chosen as reference test. The considered scenarios involve the removal of four distinct columns: two internal ones with different boundary conditions, an edge column and a corner column, all the elements are placed at the ground floor level. The third aim is to evaluate the influence of several parameters on global response, to identify their possible influence on the phenomenon and to highlight which among these have a determining impact on the structural response. The factors investigated are: primary beams depth, columns depth, presence of bracing systems, continuous reinforcement amount and seismic detailing. The achieved results provide precise information on the overall structural behaviour, highlighting the key role played by certain factors such as the percentage of continuous reinforcement in the beams and the importance of seismic detailing. At the same time the analyses have highlighted the marginal influence exerted by other parameters like the stiffening contribution given by a bracing system or stiffer columns, whose effects may be considered negligible.