Uncertainty analysis through development of seismic fragility curve for a SMRF structure using adaptive neuro-fuzzy inference system based on fuzzy c-means algorithm
| Autor: | S. B. Beheshti Aval, Fooad Karimi Ghaleh Jough |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Adaptive neuro fuzzy inference system
Mathematical optimization Monte Carlo method General Engineering 020101 civil engineering 02 engineering and technology Incremental Dynamic Analysis Fuzzy logic 0201 civil engineering 020303 mechanical engineering & transports Fragility 0203 mechanical engineering First-order second-moment method Uncertainty quantification Algorithm Uncertainty analysis Mathematics |
| Zdroj: | Scientia Iranica. |
| ISSN: | 2345-3605 |
| Popis: | The present study is focused mainly on development of the fragility curves for the sidesway collapse limit state. One important aspect of deriving fragility curves is how uncertainties are blended and incorporated into the model under seismic conditions. The collapse fragility curve is in uenced by di erent uncertainty sources. In this paper, in order to reduce the dispersion of uncertainties, Adaptive Neuro Fuzzy Inference System (ANFIS)based on the fuzzy C-means algorithm is used to derive structural collapse fragility curve, considering e ects of epistemic and aleatory uncertainties associated with seismic loads and structural modeling. This approach is applied to a Steel Moment-Resisting Frame (SMRF) structural model whose relevant uncertainties have not been yet considered by others in particular by using ANFIS method for collapse damage state. The results show the superiority of ANFIS solution in comparison with excising probabilistic methods, e.g., First- Order Second-Moment Method (FOSM) and Monte Carlo (MC)/Response Surface Method (RSM) to incorporate epistemic uncertainty in terms of reducing computational e ort and increasing calculation accuracy. As a result, it can be concluded that, in comparison with the proposed method rather than Monte Carlo method, the mean and standard deviation are increased by 2.2% and 10%, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|