Seismic fragility analysis of RC box-girder bridges based on symbolic regression method

Created by W.Langdon from gp-bibliography.bib Revision:1.8051

@Article{REZAEI:2022:istruc,

• author = "Hossein Rezaei and Panam Zarfam and Emadaldin Mohammadi Golafshani and Gholamreza Ghodrati Amiri",
• title = "Seismic fragility analysis of {RC} box-girder bridges based on symbolic regression method",
• journal = "Structures",
• volume = "38",
• pages = "306--322",
• year = "2022",
• ISSN = "2352-0124",
• DOI = "doi:10.1016/j.istruc.2021.12.058",
• URL = "https://www.sciencedirect.com/science/article/pii/S2352012421011760",
• keywords = "genetic algorithms, genetic programming, Evolutionary computation, Regression analysis, Fragility curve, Machine learning, Automatic regression, Multi-span bridge, Nonlinear time history analysis, Bridges with Unequal-Height Piers",
• abstract = "The fragility curves are the key tools in seismic risk assessment within the performance-based earthquake engineering framework. This study employs the Symbolic Regression (SR) analysis to develop seismic response prediction models and the fragility curves for concrete box-girder bridges. Geometrical, material, ground motion, and structural uncertainties were taken into account to improve the reliability of the derived mathematical models. The responses were recorded as Engineering Demand Parameters (EDPs) through a nonlinear time history analysis of the simulated bridges and used as targets for prediction algorithms. The parameters of higher significance were then identified based on the evolutionary correlation coefficient and adopted as input parameters. Accordingly, a total of 30 mathematical models were developed for the five classes of regular and irregular bridges and six EDPs. The model accuracy and effectiveness were evaluated based on various criteria. A method was then proposed to develop fragility curves based on the resulting prediction models. The results were suggestive of the effectiveness and accuracy of the prediction models despite their simplicity. Further, the resulting fragility curves were consistent with those obtained by nonlinear time history analysis and, for different EDPs, the geometrical parameters were among the highest-correlation parameters after the seismic intensity measure",

}

Genetic Programming entries for Hossein Rezaei Panam Zarfam Emadaldin Mohammadi Golafshani Gholamreza Ghodrati Amiri

Citations