ABSTRACT
To assess the reliability of existing bridges, in general, extreme effects on a structure are analyzed, which allows predicting their behavior. In particular cases of small bridges, effects caused by fatigue due to the cyclic traffic loading might be significant. The main interest of the current paper is the possibility of applying the extreme value analysis to dissimilar bridges using data from different types of monitoring. Research in this area is carried out on the Anneau bridge in Switzerland and on the Millau viaduct, France, in order to compare the fatigue behavior based on signals from strain gauges in one case and on the WIM data in the other. The estimation of the fatigue life is based on the Extreme Value Theory and, in particular, the Generalized Pareto Distribution. The results aid to derive the efficiency of the method for the predictions of the high cycle fatigue action.
