Damage detection using piezoelectric transducers and the Lamb wave approach: II. Robustand quantitative decision making

Abstract

The propagation of Lamb waves generated by piezoelectric transducers in a one-dimensionalstructure has been studied comprehensively in part I of this two-paper series.Using the information embedded in the propagating waveforms, we expect tomake a decision on whether damage has occurred; however, environmental andoperational variances inevitably complicate the problem. To better detect thedamage under these variances, we present in this paper a robust and quantitativedecision-making methodology involving advanced signal processing and statisticalanalysis. In order to statistically evaluate the features in Lamb wave propagation inthe presence of noise, we collect multiple time series (baseline signals) from theundamaged beam. A combination of the improved adaptive harmonic wavelettransform (AHWT) and the principal component analysis (PCA) is performed on thebaseline signals to highlight the critical features of Lamb wave propagation in theundamaged structure. The detection of damage is facilitated by comparing thefeatures of the test signal collected from the test structure (damaged or undamaged)with the features of the baseline signals. In this process, we employ Hotelling’sT2 statistical analysis to first purify the baseline dataset and then to quantify thedeviation of the test data vector from the baseline dataset. Through experimentaland numerical studies, we systematically investigate the proposed methodologyin terms of the detectability (capability of detecting damage), the sensitivity(with respect to damage severity and excitation frequency) and the robustnessagainst noises. The parametric studies also validate, from the signal processingstandpoint, the guidelines of Lamb-wave-based damage detection developed in part I.