Structural Damage Identification in an Aeronautical Panel Exploring the PCA and Delaunay Triangulation

Abstract

This paper addresses a novel approach for damage identification in Structural Health Monitoring (SHM) systems based on Principal Component Analysis (PCA) and Delaunay Triangulation (DT). The proposed method explores Electromechanical Impedance (EMI) principle, in the time domain, to identify structural damage in an aircraft aluminum panel. The procedure is carried out exciting the PZT (Pb-Lead Zirconate Titanate) transducers using a wide band chirp signal and measuring their time responses. The PCA is performed using structural response signals and PCA loadings are considered as input to compute the DT. Root Mean Square Deviation (RMSD) metrics are computed from the area formed by the triangles obtained from the DT. Such areas vary due to the displacement of loadings caused by changes in the structural condition for each PZT sensor. These variations have provided good evidence about the effectiveness of the proposed method applied to SHM systems.