Extraction of wavelet-based damage-sensitive features for concrete moment-resisting frames

Abstract

Listen

Translate article

This paper aims to derive conventional wavelet-based damage-sensitive features (WB-DSFs) and refined wavelet-based damage-sensitive features (rWB-DSFs) for concrete moment-resisting frames (MRFs) and detection of the best mother wavelet. For this purpose, wavelet-based and refined wavelet-based damage-sensitive features extracted from the absolute acceleration responses of building models subjected to different ground motion records were evaluated by using the continuous wavelet transform. Three concrete moment-resisting frames with four-, eight- and twelve-story building frames were selected to implement wavelet-based and refined wavelet-based damage-sensitive features. The wavelet coefficients were calculated from the acceleration signals, and different damage-sensitive features were specified by wavelet energies based on the appropriate scales and times. Then, the conventional wavelet-based damage-sensitive features were modified to address the effects of higher mode contributions. The results confirmed the efficiency of wavelet-based and refined wavelet-based damage-sensitive features for damage diagnosis and migration of these damage-sensitive features during damage progress. Furthermore, the findings of this study showed that the mother wavelet family of bior3.3 yields more accurate damage diagnosis states for reinforced concrete moment-resisting frames. The results show that the damage pattern, as well as the damaged floor can be detected using the proposed wavelet-based damage-sensitive features.