On the Fitting of Multimodal Intensity Functions to Complex Accelerograms

Abstract eng:
The estimation of the intensity function of an earthquake record is a key step for the non-stationary stochastic representation of seismic action in dynamic structural analyses. In spite of its importance, only a few procedures have been applied to obtain the intensity function of a given earthquake record. Amongst them, the underlying stationary process (USP) method, proposed by Ferrer and Sánchez-Carratalá (2006), is the only procedure that can be considered as a standard method for earthquake analysis, since it can be systematically applied to fit any prescribed intensity function to any real record. Furthermore, it allows the quantitative assessment of the fitting by using some error parameter calculated from the underlying process. The USP method has been extensively applied to many accelerograms from several seismic regions using common unimodal theoretical or code-based intensity functions. However, these unimodal functions only attempt to model the amplitude evolution over the time of seismic events caused by a singular and instantaneous rupture process. The analysis of complex accelerograms, with two or more separate peaks produced by several fault ruptures or very different wave paths, can be accomplished by the composition of two or more single-peak intensity functions. In the present paper, the standard USP method is extended to solve the problem of fitting the amplitude modulation function of multimodal accelerograms. A numerical application of the proposed method to some real multimodal earthquake records is carried out, showing the ability of the USP method to accurately obtain the amplitude modulation function of complex accelerograms.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket