Structural components made of composite materials are being used more often in aerospace and aeronautic structures due to their well-known properties such as high mass specific stiffness and strength. However, their application also increases the analysis complexity of such structures. Structural health monitoring (SHM) systems for these structures aim to determine the status of the system in real time such that a longer safe life and lower operational costs can be guaranteed. On that account, this paper is concerned with the experimental validation of a structural
health monitoring methodology where a damage detection and classification scheme based on an acousto-ultrasonic (AU) approach is applied to a composite panel incorporating stiffening elements using a piezoelectric active sensor network in conjunction with time-frequency multiresolution analysis and non-linear feature extraction. Therefore, structural dynamic responses from the simplified aircraft composite skin panel are collected and signal features are then extracted with a signal processing and data fusion methodology in terms of the wavelet transform technique and hierarchical non-linear principal component analysis. A critical comparison with linear feature extraction methods indicates that the proposed method outperforms the traditional linear methods for the purpose of damage
classification. Additionally, results show that all the damages were detectable and
classifiable, and the selected features proved capable of separating all damage
conditions from the undamaged state.
