Life-Cycle Damage and Downtime Cost Estimates for Structural Systems Subjected to Seismic Loads

Cornell University, Ithaca NY

Investigators

Abstract

The main objective of this award is the development of an accurate, efficient, and conceptually simple methodology for assessing the life-cycle seismic performance of realistic structural systems. The research is motivated by its potential to contribute to the practical implementation of performance-based seismic design concepts. A first step of the research is the development of a probabilistic model describing the seismic hazard at a particular site. The model will be calibrated with the data available from the United States Geological Survey. The model will deliver arrival times, magnitudes, and epicenter locations for the sequence of seismic events likely to occur at a particular site during a specified time interval, as well as ground motions during these events. A second step of the research is the development of reduced order models characterizing probabilistically seismic ground acceleration records at a site. Resulting reduced order model will be used to calculate system fragilities as functions of earthquake magnitudes and epicenter-to-site distances, rather than a single seismic intensity parameter. These developments will be applied to estimate life-time damage and downtime costs for realistic structural systems and construct metrics rating the seismic performance of design alternatives. The completed research will provide an accurate and conceptually simple methodology for estimating life-time damage and downtime costs for structural systems, which are essential ingredients for the developments performance-based design procedures. In addition to providing useful input to performance-based design-related studies, the project is likely to open new research directions on the analysis of complex, large scale dynamic systems with uncertain properties subjected to random actions. The reduced order models developed in this study can be used to design efficient protocols for testing equipments and other systems subjected to random excitations. These protocols will extract maximum information from a specified number of tests by selecting representative samples of these excitations. Graduate and undergraduate engineering students will benefit from results of the proposed research through classroom instruction, participation in the project as members of the research team, and software packages to be developed in the project.