Data Fusion about Vine-Copula for Bridge Section Reliability Considering Nonlinear Correlation of Failure Modes

Abstract:

Bridge section reliability analysis method is reasonably carried on with the fusion of structural health monitoring data. Firstly, the vinecopula models considering the nonlinear correlation of multiple monitored variables were established based on the extreme stress data at the multiple monitored points of bridge section, which make the extreme stress data fusion achieved. Secondly, the vine copula models considering the nonlinear correlation of failure modes about bridge section were built with the performance functions about the multiple monitored points, further, through combining the built vine copula models with first order second moment (FOSM) method, the bridge section reliability considering the nonlinear correlation of failure modes was analyzed. Finally, the monitored data of an existing bridge was provided to illustrate the proposed model and method. The results show that the obtained bridge section reliability with considering the nonlinear correlation of failure modes is bigger than that without considering the correlation of failure modes. It is illustrated that the obtained results without considering the correlation of failure modes are conservative.

Key words:bridge; girder section; nonlinear correlation; VineCopula model; first order second moment (FOSM) method; reliability analysis