研究目的
Investigating the disaggregation of seismic hazard based on the eta spectral shape indicator to refine conditional mean spectra and improve ground-motion record selection for structural analysis.
研究成果
The disaggregation of the eta indicator refines the conditional mean spectra, particularly in the short period range, leading to more accurate predictions of structural performance and annual probability of failure. The proposed method outperforms existing approximations and is effective for ground-motion selection with reduced records.
研究不足
The results are case-specific to the single ideal site with a linear seismic source considered; they may not generalize to other site conditions or complex seismic sources. The study relies on approximations and assumptions in GMPEs and disaggregation, which could introduce uncertainties.
1：Experimental Design and Method Selection:
The study uses probabilistic seismic hazard analysis (PSHA) and disaggregation to analyze the eta indicator. A simplified linear seismic source model is assumed for an ideal site. Methods include linear regression, correlation models, and genetic algorithms for ground-motion selection.
2：Sample Selection and Data Sources:
An ideal site with a single line source of length 30 km, located 15 km away, is used. Ground-motion records are sourced from the PEER NGA database, specifically 44 records from various earthquakes.
3：List of Experimental Equipment and Materials:
No specific physical equipment is mentioned; the study relies on computational methods and software for PSHA, disaggregation, and structural analysis (e.g., incremental dynamic analysis).
4：Experimental Procedures and Operational Workflow:
Steps include performing PSHA to obtain hazard curves, disaggregating seismic hazard for epsilon and eta, calculating conditional mean spectra (CMS and ECMS), selecting ground motions using genetic algorithms, and analyzing a three-storey reinforced concrete building using nonlinear response history analysis to compute the annual probability of failure.
5：Data Analysis Methods:
Statistical methods include linear regression, correlation coefficients, and error calculations. Software tools are implied for PSHA and structural analysis, but not specified.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
