Spectrum-to-spectrum methods for the generation of elastic floor acceleration spectra

In seismic codes, the acceleration demand of nonstructural components is commonly expressed in terms of floor response spectra and estimated by means of simple predictive equations. By using the latter, response-history analysis of the structure is not required, being floor spectra calculated directly from the peak ground acceleration expected at the site. The price for this simplicity in the method used for the estimation of floor spectra is the generally poor approximation of the obtained predictions. Codes’ equations, in fact, do not explicitly account for important factors influencing floor spectra, such as the contribution of the higher modes of vibration of the structure and the actual value of the nonstructural components’ damping ratio. Alternative spectrum-to-spectrum methods for direct generation of floor spectra have been proposed, which include these factors and improve the accuracy of the predictions. Different approaches have been used and several methods developed. Despite large research effort, however, a comparative evaluation of the currently available proposals is still lacking. The objective of this paper is to fill this gap, by reviewing selected proposals representative of practice-oriented spectrum-to-spectrum methods. A case study consisting in a six-story frame is analyzed and predictions obtained with the investigated methods are compared with exact floor spectra derived from time-history analyses of the structure, as well as spectra calculated using the Eurocode 8 equation.