STRUCTURE-SPECIFIC SELECTION OF EARTHQUAKE STRONG GROUND MOTIONS FOR NONLINEAR ANALYSIS OF RC STRUCTURE

Response history analysis (RHA) constitutes nowadays the most prevailing method to undertake linear and nonlinear analysis of structural systems being subjected to the time-varying earthquake forces. Compared to the conventional analysis methods (i.e., modal response spectrum and equivalent static analysis respectively), the superiority of the RHA is mainly associated with the capacity to: (i) identify the hierarchy of the failure mechanisms, (ii) account for the energy dissipation and force redistribution mechanisms as well as, (iii) favor the control of both the structural and non structural damage during the strong ground shaking. It is a time-domain structural analysis method and as such, demands the use of suite of earthquake motions that correspond to a predefined earthquake scenario. Intensive research effort has been lately spent on scrutinizing methods for selecting and scaling earthquake strong ground motions since it has been shown that the earthquake excitations and their inherent uncertainty can affect drastically the calculated response and lead, for example, to highly scattered structural analysis results. The latter undermines, though, the reliability of the structural demand parameters, which is of high relevance mainly for design and design verification purposes. Along these lines, a structure-specific method for selecting and scaling earthquake records has been recently introduced accounting explicitly, among other factors, for the dynamic properties of the structure under study and prioritizing sets of motions that disfavor the structural response variability. This method, being developed under the ISSARS computational framework, has been successfully tested in terms of linear RHA results. To further extend the validation and hence, the applicability of the proposed method, its performance is evaluated via the current study in terms of nonlinear analysis. Especially, a rc multistory, frame-resisting structure is modelled by using OpenSees while multiple nonlinear RHA are undertaken by the use of earthquake motions, the latter being appropriately selected, formed into suites and prioritized for structural analysis purposes via the use of the structure-specific earthquake records selection method. The code-based, conventional method to select and form sets of earthquake motions is also applied and the intra-suite variability of the corresponding response results is compared with the one induced by the structure specific earthquake records selection method. The latter is expected to lead to response parameters with lower variability and hence, design values of increased reliability.