Abstract
Structural health monitoring (SHM) based on electromechanical impedance (EMI) measurements has been extensively studied in recent years and stands out as an effective damage detection approach. The EMI method commonly uses low-cost piezoelectric transducers and structural diagnosis can be accomplished by analyzing the changes in the impedance measurements from the transducers attached to the host structure. Although many studies have proved the effectiveness of the EMI method, one of the most reported problems is the variations in the impedance caused by external noise, which can alter the impedance measurements and limit the effectiveness of the detection and quantification of damage using conventional basic damage indices. Therefore, this paper proposes a new approach for structure feature extraction in noisy environments based on the chromatic signal processing technique associated with the Hinkley criterion, thus aiming to expand the applicability of the EMI method to real applications. Experimental tests were carried out on aluminum and carbon fiber composite plates subjected to various signal noise levels. The experimental results indicate that the proposed approach is more effective for damage detection compared to the basic indices commonly reported in the literature.