Structural Damage Diagnosis of Reinforced Concrete Moment-Resisting Frame using Wavelet Based-Damage Sensitive Feature

This paper describes the derivation of a wavelet-based damage-sensitive feature for structural damage diagnosis in a reinforced concrete moment-resisting frame. The wavelet-based damage-sensitive feature is based on the wavelet transform of the time-history acceleration response, obtained from incremental dynamic analysis of the floor absolute acceleration response. By applying mathematical relationships to the wavelet coefficients, the wavelet-based damage-sensitive feature is extracted as the energy of the wavelet coefficients at appropriate scales. Furthermore, the theoretical correlation between the output of wavelet analysis and dynamic structural behaviors that play a vital role in damage diagnosis methods was derived. The effectiveness of the wavelet-based damage-sensitive feature is evaluated by simulating a 4-story reinforced concrete frame subjected to various ground motion records using the continuous bior3.3 mother wavelet family. The wavelet-based damage-sensitive feature measures changes in wavelet energy at the natural frequency of the undamaged structure as damage progresses, and the results show that wavelet-based damage-sensitive feature values gradually decrease as damage increases.