Grading diagnosis of beam bridge support separation disease within one cluster based on the difference in the probability density function of the measured data

Abstract

Bridge supports are the main bridge devices for resisting loads; thus, if separation disease occurs in bridge supports, the structural safety of a bridge cannot be ensured. There are very few studies using measured data for diagnosing the separation disease of support, and more studies are urgently needed. To address the abovementioned issue, a method of grading diagnosis of separation disease of beam bridge supports within one cluster based on the difference in the probability density function of measured data is proposed in this study. By making full use of the symmetry of the bridge structure and the rule of the beam body morphologic change with the state of the support, the instantaneous diagnosis index and the cumulative diagnosis index are constructed by utilizing the relationship of angular displacement at supports within one cluster. A grading diagnosis for the separation disease of support with the data measured by a wireless sensing network is proposed, which indirectly solves the difficulty of detecting the separation disease by directly measuring the supporting reaction force. The proposed method is validated by using both numerical examples and measured data from an actual bridge. The results show that the proposed method is effective for diagnosing the separation disease of bridge supports and is suitable for application to actual bridges.