Damage Categorization of Glass/Epoxy Composite Material Under Mode II Delamination Using Acoustic Emission Data: A Clustering Approach to Elucidate Wavelet Transformation Analysis

Delamination is the most common failure mode in composite materials, since it will result in the reduction of stiffness and can grow throughout other layers. Delamination consists of two main stages: initiation and propagation. Understanding the behavior of the material in these zones is imperative, hence the identification of this mode of failure is of great importance. There are several methods to identify damage in materials, one of which is using acoustic emission (AE) signals. Most of the pervious works have used statistical methods based on the energy of AE signals, but in this study, the normalized form of shock wave is used. The aim of this study is to extract a general pattern for specific damage from AE signals including all of the other damage signals. The method consists of a discrete wavelet packet decomposition of AE signals accompanied with a clustering algorithm, which gives the distribution of the normalized AE signal energy on the frequency band. Test set-up involved End Notched Flexure (ENF) test to detect mode II delamination on glass/epoxy composite material. The data obtained from ENF test specimens is used for the wavelet packet decomposition, and the energy of different levels of decomposition for each shock wave is clustered using different clustering algorithms including K-means and Fuzzy C-mean. Scanning Electron Microscope was used to validate the results.