Damage detection in thin plates and aerospace structures with the electromechanical impedance method. Struct Health Monit 2005; 4(2): 99–118

The use of the electro-mechanical (E/M) impedance method for health monitoring of thin plates and aerospace structures is described. As a nondestructive evaluation technology, the E/M impedance method allows us to identify the local dynamics of the structure directly through the E/M impedance signatures of piezoelectric wafer active sensors (PWAS) permanently mounted to the structure. An analytical model for 2-D thin-wall structures, which predicts the E/M impedance response at PWAS terminals, was developed and validated. The model accounts for axial and flexural vibrations of the structure and considers both the structural dynamics and the sensor dynamics. Calibration experiments performed on circular thin plates with centrally attached PWAS showed that the presence of damage modifies the high-frequency E/M impedance spectrum causing frequency shifts, peak splitting, and appearance of new harmonics. Overall-statistics damage metrics and probabilistic neural network (PNN) are used to classify the spectral data and identify damage severity. On thin-wall aircraft panels, the presence of damage influences the sensors E/M impedance spectrum.When crack damage is in the PWAS medium field, changes in the distribution of harmonics take place and when crack damage is in the PWAS near field, changes in both the harmonics and the dereverberated response are observed. These effects are successfully classified with PNN and overall-statistics metrics, respectively. This proves that permanently attached PWAS in conjunction with the E/M impedance method can be successfully used in structural health monitoring to detect the presence of incipient damage through the examination and classification of the high-frequency E/M impedance spectra. Keywords electro-mechanical impedance piezoelectric wafer active sensors vibration of circular plates aging aircraft structural health monitoring overall statistics neural networks E/M impedance PWAS EMI SHM IVHM PNN