Quantifying the impact of normalized period on seismic demand model of ductile columns under pulse-like ground motions

Near field ground motion is found to be able to induce a larger seismic demand on structures, as concluded from velocity pulse observed by various researchers in the past. The pulse period (Tp) which can characterize velocity pulse is proven to have a significant influence on structural responses. This paper, therefore, develops a framework that can be used to establish the curvature ductility demand spectra that describe the relationship between the ratio of the structural fundamental period (T1) to Tp and the corresponding seismic demand based on the Artificial Neural Network (ANN) model. The effect of structural and seismic parameters on seismic demand spectra can be well captured by the defined spectra parameters including Dmax (peak ductility demand), Dmin (minimum ductility demand), Dδ (initial ductility demand), µT (most unfavorable location of T1/Tp) and sensitive interval (most unfavorable design range of T1/Tp), which make the spectra more practical for seismic response analysis. The proposed strategy is then applied to the analysis of reinforced concrete (RC) columns under the action of seismic loading where the influence of longitudinal reinforcement ratio (ρ) and damping ratio (ς) is systematically investigated and compared.