A comparison of signal processing techniques for impedance-based damage characterization in carbon fibers under noisy inspections

Among several structural health monitoring (SHM) systems, the electro-mechanical impedance (EMI) method is an effective failure detection approach in composite components. In the EMI method, damage detection is achieved by comparing the behaviour of piezoelectric impedance signatures, which are related mathematically to the mechanical impedance of the host structure. Although many studies have proved the effectiveness of the EMI method on composites, many practical issues such signal changes caused by noise and vibrations have limited its use in real-life applications. This work presents an experimental analysis of white noise effects on the impedance monitoring of a carbon fibre reinforced plastic composite plate, by applying the correlation coefficient deviation metric (CCDM) and cross-correlation square deviation (CCSD) index on the response signal of a pitch-catch mode piezoelectric transducer. Experimental results revealed that the CCSD was more sensitive to damage detection under noisy inspections.