Optimal Ground Motion Scaling under Spectral Variation Constraints

Abstract

In seismic design, scaling of ground motion records is critical for the accuracy of structural analysis. According to the Turkish Building Earthquake Code (TBEC 2018), for time history analyses, actual ground motions can be scaled according to the design spectrum. The selection of scale factors by the designer can distort the characteristics of the records and lead to erroneous structural analyses. Especially high scale factors can distort the characteristics of the original record. This study proposes a multi-criteria optimization process for determining scaling factors. The objective function was defined as minimizing the impact on the characteristics of the scaled acceleration records. Harmony Search Algorithm (HSA) and Grey Wolf Algorithm (GWO) were used in the optimization process. Additionally, the scaling process was carried out using the SeismoMatch program. The obtained results were compared in terms of the accelerograms and characteristics of the records. The effects of the optimum scale factors were also examined by conducting time-history analyses on a seven story reinforced concrete structure. According to the optimized results, the peak acceleration value in the accelerograms increased by 1.8 times, while scaling with SeismoMatch resulted in a 7.5 fold increase. Following an optimization process conducted on 11 different ground motion records, it has been demonstrated that the proposed variation-constrained method fully satisfies the TBEC 2018 criteria while exhibiting proficiency in preserving the seismological characteristics and energy content of the acceleration records.