An efficient algorithm to simulate site‐based ground motions that match a target spectrum

Abstract

AbstractAn efficient algorithm is presented to simulate site‐based ground motions using Dabaghi and Der Kiureghian (2018) and Rezaeian and Der Kiureghian (2012) that achieve target ground motion intensity measures between a range of periods. The proposed algorithm uses predictive relations between the model parameters (MP) and the RotD50 spectral acceleration spectrum, RotD50Sa,Tar(T), of the site‐based ground motions. A correlation structure is estimated to account for the cross‐spectral correlations between the periods. The algorithm requires a set of five seismic event parameters (including Magnitude, Rupture Distance, Soil Shear‐Wave Velocity, Fault Mechanism, Fault Geometry) and the target spectrum (including the range of periods whose corresponding spectral accelerations are required to be matched). The proposed algorithm uses event parameters to generate appropriate sets of model parameters. Using the developed predictive relations and correlation structure, the algorithm then carefully selects the model parameters that are statistically likely to generate a ground motion waveform possessing the required target spectrum between the inputted range of periods. Unlike the traditional scaling methods, this algorithm does not alter the amplitude, time‐ and frequency‐domain characteristics of the ground motions; hence, it leads to realistic simulated ground motions. The algorithm is used to generate ground motions for response assessment of a two‐span ordinary box‐girder bridge structure, and the results are compared against the traditional methods of selecting and scaling recorded ground motions. This demonstration indicates that the conventional methods of selecting and scaling recorded ground motions lead to a bias in the bridge's responses and the proposed algorithm leads to statistical consistency in the results.