Development of Seismic Fragility Curves of Horizontal Curved Bridge Using Neural Network Prediction.

Abstract eng:
In this study a Soft Computing approach for the seismic fragility assessment of horizontal curved bridge is developed. In recent years, seismic fragility curves are often determined by using analytical method in structures. In order to build neural network structure, Nonlinear time history analysis is performed by 129 natural records in OpenSees. This records have been chosen from the PEER strong motion database and scaled on 0.1g to 1.3g. The structure of the neural network is based on input ground motions and output of nonlinear dynamic analyses of bridge. Arias Intensity, cumulative absolute velocity, characteristic intensity and specific energy density reflect the amplitude, the duration of a strong ground motion, the frequency content and energy respectively, and they correlate well with structural damage. Kolmogorov-Smirnov and Shapiro-Wilk tests was performed to normalize the data. Approach to reducing the computational effort in the evaluation of fragility, a neural network was considered in this study, which can provide accurate predictions of the structural response. The proposed approach is applied for bridge and a reduction of magnitude is achieved in the computational effort.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket