Evaluation the effect of amplitude scaling of real ground motions on seismic demands accounting different structural characteristics and soil classes

Abstract

Selection and scaling of input ground motion records for nonlinear dynamic analysis of structures has become an important issue and studies on this subject can provide valuable information for earthquake engineering practice. In this study, the effect of different amplitude scaling approaches and scaling levels based on spectrum-compatible selection were investigated for various local soil classes. A recent Turkish Building Earthquake Code was considered and simple structural models that reflected the different characteristics of structures were employed. The dynamic attributes of the models were represented by using different lateral strength ratios, post yield stiffness ratios and natural vibration periods. The maximum displacement, residual displacement, and input energy demands were used to evaluate the efficiency of scaling approaches and levels. In addition, some of the commonly used intensity measures that reflected the different characteristics of ground motions were considered for the evaluation. Significance of the uncertainties associated with structural models and their impact on the results were also statistically evaluated by variance analysis. Moreover, a probability-based method was adapted and influence of scaling approaches and their dependence to local soil types was established by examining the statistical mean of maximum displacement demands in terms of amplification factors. Results demonstrated that distribution of mean of the structural responses and the intensity measures are free from scaling approaches and scaling levels, and the differences between the demands obtained for different scaling levels are statistically insignificant for 90% confidence level. Probability-based evaluations have also revealed that amplification factors for seismic demands can be assumed as independent from scaling approaches, scaling levels and local soil classes. © 2023, The Author(s), under exclusive licence to Springer Nature B.V.