MODELLING AND SEISMIC PERFORMANCE EVALUATION OF INDIAN RC FRAME BUILDINGS

Abstract

Modelling of RC frame buildings to simulate their seismic behaviour is a complex task. The task is made further complicated by presence of various deficiencies and irregularities present in existing buildings. In the present study, first a review of the current Indian practices on modelling, design and construction of RC buildings is performed. The current design standards are reviewed and a survey of multi-storey RC buildings in a model township within the national capital region is performed. Common irregularities and deficiencies in Indian construction and the major issues related to modelling and simulation of seismic response of RC buildings are identified. Estimation of effective stiffness of members is a crucial issue in simulation of seismic behaviour of RC buildings, as the concrete cracks and steel yields and slips under seismic action. Different parameters influencing the effective stiffness of RC members are identified and their relative influence is studied using the three component approach. Based on the parametric study, lower-bound and upper-bound estimates for effective stiffness are obtained and compared with the experimental results compiled in PEER Structural Performance Database. Different stiffness relationships available in the literature are compared with the obtained lower-bound and upper-bound estimates and new relationships for normal and high-strength concrete frame members are proposed. The relationships have been developed to represent average conditions in practical range of building frame members to avoid cumbersome calculations in detailed analysis of individual members. in The influence of effective member stiffness assumptions on estimated seismic performance of RC frame buildings is studied using Nonlinear Static Procedure. First the influence of effective stiffness on distribution of member forces and yield pattern is studied. It is shown that not only the stiffness and displacement of the building is affected by the variation in member stiffness but the predicted strength is also influenced due to alteration of yield pattern. The variation of axial load and the consequent variation of effective stiffness with lateral displacement in buildings designed for gravity only and in buildings designed for combined effect of gravity and earthquake are also studied. It is observed that the effective stiffness of columns depends on the applied moment and hence varies with lateral displacement. As most of the columns, contrary to beams, are not expected to yield during earthquake, the real stiffness is much different than that estimated at yield of columns. It is quite cumbersome to consider the variation of stiffness with axial load and moment during the design, and hence the design codes provide simplifying assumptions for stiffness of columns. The codes generally recommend column stiffness corresponding to yield at axial load due to gravity. The influence of effective stiffness provisions of different codes on estimated nonlinear behaviour is studied for a typical eight storey building and it is found that the predicted response may be drastically different than the realistic response. It is a common practice to construct the RC floor/roof slabs monolithically with the beams. This makes portions of slab to act like flanges to the beams and enhance their stiffness and strength. The effect of flange action of slab is first studied at member level and then at building level. Multiplication factors are proposed to obtain the effective stiffness and strength of flanged beams from the corresponding web

sections. Influence of flange action of slab on estimated seismic response of RC buildings with different design and detailing is also studied. Seismic performance and vulnerability of RC frame buildings with the most common design and detailing deficiencies are estimated. The deterministic analysis is not able to account for the uncertainty associated with capacity, damage states and ground shaking. Fragility analysis, on the other hand, provides a more realistic estimate of expected damage and associated risk. Fragility curves and DPMs for the RC buildings with design and detailing deficiencies are obtained using HAZUS approach and compared with the properly designed and constructed buildings. It is shown that buildings designed for gravity load alone have some overstrength and can withstand earthquakes of moderate intensity, provided these are detailed and constructed properly. On the other hand, the buildings designed for earthquake forces but with improper detailing and anchorage of reinforcement are expected to perform much poorly.