Fusion of Wavelet Packets and Neural Network in Detection of Composites
Yaojun Wu, Xizhi Shi, Tian Zhuang · AIAA Journal · 2000
A new data-fusion method is proposed for the damage detection of anisotropic composite materials. Based on signal processing theory, we combine wavelet packets, which can decompose signals into a tiling of plane of the time frequency and the feature information in different frequency bands, with autoregression spectrum analysis to extract features and recognize the characteristic signals sampled at experiments of damage detection of composites by vibration. These features are fed into the wavelet neural network as the input patterns for training and classifying. Analysis on the signals obtained in the damage detection experiment of composites demonstrates the effectiveness of the proposed method.