Damage features extraction of prestressed near-surface mounted CFRP beams based on tunable Q-factor wavelet transform and improved variational modal decomposition

Damage detection of prestressed near-surface mounted (NSM) carbon fiber reinforced polymer (CFRP) beams has attracted much attention over the last few decades due to the dramatic increase in the proportion of prestressed NSM CFRP beams in the civil infrastructure. Although many studies have demonstrated the effectiveness of piezoelectric transducer-based damage detection techniques, one of the most reported problems is the variation in signal vibration characteristics caused by external noise, which limits the effectiveness of detection and quantification using traditional basic damage indicators. This study proposed a novel signal processing method of damage feature extraction in a noisy environment based on the tunable Q-factor wavelet transform with improved variational modal decomposition, which is aimed to expand the applicability of the piezoelectric transducer-based damage detection techniques in the damage diagnosis of the prestressed NSM CFRP beam. The proposed method is validated by a numerical case study of an analog signal and an experimental study of NSM CFRP beams. The results show that the tunable Q-factor wavelet transform and improved variational modal decomposition can be employed to remove the environment noise and extract the damage feature of the NSM CFRP beam, which helps in the damage assessment of the NSM CFRP beam.