An On-Board Damage Detection Method for Heavy-Haul Railway Bridge Based on Sensitivity Analysis of Bogie Responses

The on-board damage detection methods are recently applied to the structural health monitoring (SHM) for railway bridge, and with more economical and efficient than the traditional methods. However, to improve the calculation efficiency and avoid the detection sensitivity of the on-board methods decreasing, most researchers always leave out the track structure in simulation models, while it is not consistent with the complex actual situation. To realize the accurate detection, a novel on-board damage detection method for heavy-haul railway bridge using bogie accelerations is put forward, which is availably implemented for both of damage localization and damage quantification. A relatively complete model considering train–track–bridge interaction is established, and a model simplification method of railway bridge is proposed based on the finite element shape function to improve calculation efficiency further. The sensitivity of the bogie vertical accelerations to damage is analyzed to construct the sensitivity equation of coupling system. By iteratively solving the sensitivity equation and modifying the numerical model, the bridge damage can be identified. For an exhaustive evaluation of the feasibility of the proposed method, different damage scenarios are taken into account. The results demonstrate that the proposed method has a better performance in localizing and quantifying damage than the traditional method. Moreover, the influences of operational parameters such as track irregularity and vehicle speed on the validity and robustness of method are investigated. It is found that the rougher track irregularity and the higher vehicle speed seem to be more in favor of damage identification.