05-05-2011, 09:41 AM
ABSTRACT:
The Pushover procedure is extended for seismic damageassessment of asymmetrical buildings. By mean of an example, it is shownthat the accuracy of the proposed 3-D pushover analysis is similar to thoseapplied to planar structures. The procedure is found to be more successfulin estimating the global response parameters such as interstorey driftsthan local damage indicators such as beam or column ductility demands.Keywords: Pushover analysis; Nonlinear analysis; Dynamic analysis,Multistorey building; 3-D analysis
1. Introduction
Recent interests in the development of performance basedcodes for the design or rehabilitation of buildings inseismic active areas show that an inelastic procedurecommonly referred to as the pushover analysis is a viablemethod to assess damage vulnerability of buildings [1, 9].In brief, a pushover analysis is a series of incremental staticanalyses carried out to develop a capacity curve for thebuilding. Based on the capacity curve, a target displacementwhich is an estimate of the displacement that thedesign earthquake will produce on the building isdetermined. The extent of damage experienced by thebuilding at this target displacement is consideredrepresentative of the damage experienced by the buildingwhen subjected to design level ground shaking.This approach has been developed by many researchers[3, 4, 7, 8], with minor variation in computationprocedure. In most studies, the method was applied tosymmetrical structures. Assuming the floors act as rigiddiaphragms, the state of damage of the building can beinferred from applying a two dimensional pushoveranalysis on the building. If all the lateral load-resistingelements are similar, one can further simplify the problemto perform pushover analyses on a typical element of thebuilding. The advantages and the limitations of 2Dpushover analyses for damage assessment are describedby Lawson et al [5]. One limitation is that the method doesnot account for the three-dimensional effect [9].This paper extends the pushover analysis to coverplan-eccentric buildings and take the three-dimensionaltorsional effect into account. Because of torsionaldeformation, floor displacements of the building willconsist of both translational and rotational components.The lateral load resisting elements located at differentpositions in plan will experience different deformations.Torsional effect can be particularly damaging to elementslocated at or near the flexible edge of the building wherethe translational and rotational components of the floordisplacement are additive. In view of the damage observedin many eccentric buildings in past earthquakes, it is thepurpose of the present study to extend the 2-D pushoveranalysis procedure so that the vulnerability of elementslocated near the flexible edge of plan-eccentric buildingscan be assessed.
2. Procedure
The procedure in the 3-D pushover analysis followssimilar steps used in the two dimensional pushoveranalysis. First, the capacity curve is obtained by performinga series of three dimensional static analyses on thebuilding when it is subjected to a set of forces V{f}applied at the centres of mass (CM) of the floors of thebuilding. V represents the base shear and {f} is thenormalised load vector. The capacity curve is given by theV-relation obtained from the static analyses where isthe CM displacement at the roof. Since the CM of thefloors may not coincide with the centres of rigidity (CR) ofthe floors, the CM roof displacement would reflect boththe translational and torsional deformation of the buildingunder seismic lateral loading. The capacity curve has aninitial linear range with a slope k, and can be expressed inthe form V= k x G () where G () is a function describingthe shape of the capacity curve. An equivalent singledegree of freedom (SDOF) system can be established toobtain the target displacement as follows
Download full report
http://www.iiees.ac.ir/english/images/st...er00_3.pdf