Search

Various structural failure modes exhibit great differences in occurrence probabilities, failure consequences and intrinsic characteristics, etc. Utilising these differences can enhance the effectiveness and efficiency of damage identification research, which has currently received little coverage. Therefore, a new idea, developing mode-specific damage detection methods, is suggested in the paper, with validation on beam structures. Firstly, structural behaviour was investigated from the perspective of stochastic traits and the structure design theory. It was concluded that developing damage detection methods for probabilistically dominant modes had potential to enhance the effectiveness of the identification significantly. Secondly, the strategy was applied to simply supported beams for the purpose of verification. When following the flexural mode, a major failure pattern of the beams, the structures demonstrated a very interesting feature – linearized deflection curves. Hence, damage severity and location indices were defined and an analytical model was built by quantifying the feature and then linking the indices to the curvature of the beams. Thirdly, experiments on reinforced concrete (RC) beams were conducted. The results showed that both the flexural mode and the characteristic of the linearized deflection curves were observed on all specimens; that there was a strong correlation between the damage severity index and the state of the defects on the beams; that the location indices corresponding to damaged and intact sections varied completely differently in quality and quantity. Thus, implementing the mode-specific strategy improves the specificity and efficiency of the damage identification dramatically. [ABSTRACT FROM AUTHOR]