Damage identification in cable-stayed bridges based on the redistribution of dead and thermal loads

Based on the response of a bridge to various excitations, Structural Health Monitoring (SHM) systems can provide comprehensive, real-time insight into their condition. While collected data usually represents the response to a combination of known and unknown loads, modified data, cleaned of the effects of dynamic loadings, can be made to represent the response of the structure to only constant dead loads and slowly changing and easily measured cyclic thermal loads. While it is often common to separate the effect of these two loads, the proposed method suggests that it is advantageous to consider the combined effect. Furthermore, once a limited amount of response data due to these loads has been collected, their combined effect can be easily predicted.A simplified model, consisting of the key structural components of a cable-stayed bridge, a beam, an inclined cable, and a bearing, and a more complex, that of a Finite Element cable-stayed bridge model, were used to evaluate the ability of the proposed method for identifying changes in these three elements. Beam damage was simulated as a flexural stiffness reduction for part of the beam, cable damage as an axial stiffness reduction of the cable and the bearing damage as a stiffness increase of a horizontal spring supporting the free to roll end of the beam. Using the model, it was shown that the behavior of two parameters, the slope and the intercept taken from the beam strain-temperature curve, can be used to not only identify change/damage to these members, but more importantly, determine the nature of the change/damage. Combined with the fact that the method has limited requirements for collecting, storing, and processing data, the proposed meth-odology represents a very promising Damage Identification tool.

Automated summary

Structural Health Monitoring (SHM) systems can provide real-time insight into the condition of a bridge based on its response to various excitations. The proposed method suggests that considering the combined effect of constant dead loads and cyclic thermal loads is advantageous. By collecting a limited amount of response data due to these loads, their combined effect can be easily predicted.