PROBABILISTIC SEISMIC DEMAND MODELS FOR MULTI-SPAN HIGHWAY OVERPASS BRIDGES

Abstract eng:
Probabilistic Seismic Demand Models (PSDMs) are an important component of the PEER (Pacific Earthquake Engineering Research Center) probabilistic performance-based seismic design framework. A PSDM defines a relationship between a seismic hazard Intensity Measure (IM), such as spectral displacement at the fundamental period of the bridge, and a structure-specific Engineering Demand Parameter (EDP), such as maximum column curvature. It provides a probability of exceeding an EDP value, given a value of associated IM. PSDMs are computed using a probabilistic seismic demand analysis procedure. This paper presents PSDMs developed for multi-span highway overpass bridges typical for California. These overpass bridges have two, three or four spans, circular columns with integral pile shafts, and continuous box girder superstructures. Finite element models of different configurations for each span type were developed in OpenSees. Parameters of these base models were then varied to cover a range of typical bridge designs. Each parametric bridge model was subjected to 80 ground motions recorded in California, and non-linear time-history response computed. A statistical analysis of the computed responses was used to develop a suite of bridge PSDMs. In this paper PSDMs that demonstrate commonality with optimal PSDMs derived for single-bent bridges are presented. The influence of multiple bridge spans on their seismic behavior, such as the effect of higher mode response, is also investigated. PSDMs help to illustrate the effect of variation of span length and span number on bridge performance and give designers a tool to optimize the layout of the bridge for the expected level of seismic performance.

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