Development of Intensity Compatible Time-Histories for Dynamic Analysis of Buildings

Abstract eng:
The functionality of critical infrastructures frequently depends on the availability of external support functions. Nuclear power plants are a typical example for such critical infrastructures as has been shown by the Fukushima accident. The loss of offsite power (failure of external infrastructures) caused by the earthquake triggered the accident by requesting a long term operation of emergency power supply systems that subsequently failed due to the additional tsunami impact. Therefore a realistic estimate of the damaging impact of earthquakes is crucial for preparing mitigative measures for disaster prevention and the analyses have to encompass not only the critical infrastructure itself but also external infrastructures that are necessary for safe long term operation in an emergency. The only seismological parameter that implicitly contains all required information on the damaging impact of earthquakes is intensity. Additionally, intensity information directly captures spatial variation of ground motions related to the same or similar degree damage by construction of isoseismal lines. Therefore intensities are very suitable for predicting possible impacts of earthquakes on critical infrastructures including supporting infrastructures in closed proximity that are required for their functionality. For engineering applications intensity relevant information has to be converted into engineering characteristics. Because dynamic analyses (time-history analysis, frequently even non-linear ones) became a standard approach for the design and for the validation of safety of critical infrastructures it is reasonable to develop intensity compatible time-histories for engineering application. In the paper the feasibility of a damage consistent performance-based seismic design of structures, systems and components is demonstrated. For this purpose a set of intensity compatible time-histories covering the intensity range between intensity VI and VII (EMS-98 scale) from a database of recorded time-histories is developed. These time-histories were used for the dynamic analysis of structures to develop in-structure floor response spectra. The results are compared with similar analysis results obtained by using time-histories derived from spectral matching of a uniform hazard spectrum from a large scale PSHA (PEGASOS). For the PEGASOS study the site intensity corresponding to the hazard background is known from hazard disaggregation. The comparison performed demonstrates that the use of traditional response spectrum analysis methods for the evaluation of the consequences of seismic hazard estimates derived from a PSHA in countries with low to moderate seismicity leads to conservative results. This is caused inherently by the PSHA-method. The computation of a uniform hazard spectrum (UHS) is based on the summation of exceedances probabilities of ground motion levels independently from the real damaging effects of the underlying earthquake. For a realistic assessment of the consequences of hazard estimates from a PSHA it is recommended to move towards a non-linear analysis of structures.

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