Bridge Damage Detection and Quantification Under Environmental Effects by Principal Component Analysis

Monitoring structural damage is widely used for sustaining and preserving the service life in civil structures, especially in bridges. The influence of environmental variability like temperature affects the dynamic behavior, which can mask subtler structural changes caused by damage. The direct application of vibration-based damage detection methods to measured responses without a prior treatment of the ambient data may lead to false condition assessments. In this article, the main objective is to separate the structural damage conditions from the changes caused by the environmental effects in a numerical benchmark bridge. The Principal Component Analysis (PCA) is applied to decide if the change in vibration characteristics is due to environmental effects or structural damages. The proposed approach in the use of PCA not only allows to detect the damage without the requirement of the baseline to consist of damage sensitivity features obtained from a wide range of environmental conditions, but also serves as a measure for its quantification. The effectiveness and robustness of the proposed methodology is applied to a benchmark bridge structure generated as part of COST Action TU1402 on quantifying the value of information (VoI) in SHM. The benchmark model consisted of a two-span steel bridge under environmental effects, in which two levels of damage were introduced.