Experimental Investigation of Seismic Uncertainty Propagation through Shake Table Tests

Abstract

The propagation of uncertainty in seismic response analysis is the mechanism through which ground motion (GM) uncertainty and structural parameter (SP) uncertainty contribute to response variability, and is critical for accurately quantifying seismic risk in performance-based earthquake engineering (PBEE). Response uncertainty in existing PBEE analyses is estimated either through statistical variance combination approaches or through computationally intensive time history simulations. In general, experimental exploration of the uncertainty propagation is cost prohibitive and impractical. In this paper, a series of shake table tests are conducted to generate statistical seismic response data of 12 nominally identical simple specimens subjected to a suite of 7 earthquakes at different seismic intensity levels. The test data are analyzed in order to decouple the contributions of GM uncertainty and structural parameter uncertainty on response uncertainty with different intensity levels and verifying several basic characteristics about uncertainty propagation. In addition, the accuracy of the square root sum of squares (SRSS), which is a simplified method used to combine different uncertainty sources to calculate the total response uncertainty, is evaluated through comparison with test data.