Skip to main navigation Skip to search Skip to main content

Assessment of spectrum-matched time histories for floor response spectra generation

  • Bo Li*
  • , Yang Zhou
  • *Corresponding author for this work
  • Harbin Institute of Technology Shenzhen
  • Tetra Tech

Research output: Contribution to journalArticlepeer-review

Abstract

The floor response spectra (FRS) are usually determined by time history analyses using spectrum-matched time histories. However, previous studies reported that spectrum-matched time histories may cause a biased estimate of FRS, which raises a challenge of FRS generation for nuclear power plant structures. The study focuses on resolving the challenge. We begin the study by discussing characteristics of amplitude-scaled and spectrum-matched ground motions for seismic design and analysis of nuclear facilities. Then, we explore the influence of spectral matching on the change of ground motion parameters. Using suites of amplitude-scaled and spectrum-matched ground motions, we determine FRS for examples of typical structures, and investigate how the change of ground motion parameters due to spectral matching affects the biased estimate of FRS. The investigation shows that, the change of damping scaling factor is remarkably correlated with the biased estimate of FRS. Thus, we propose the improved spectrum-matched time histories for floor response spectra generation. Taken a service building of nuclear power plants as an example, the study shows that, the improved spectrum-matched are adequate to achieve approximate unbiased estimate of FRS under different damping values.

Original languageEnglish
Pages (from-to)123-138
Number of pages16
JournalStructures
Volume38
DOIs
StatePublished - Apr 2022
Externally publishedYes

Keywords

  • Biased estimate
  • Damping scaling factor
  • Floor response spectrum
  • Spectrum-matched ground motion

Fingerprint

Dive into the research topics of 'Assessment of spectrum-matched time histories for floor response spectra generation'. Together they form a unique fingerprint.

Cite this