Locating Low Velocity Impacts on a Composite Plate Using Multi-Frequency Image Fusion and Artificial Neural Network

To find out impact induced defects in a large aerospace structure is quite time consuming due to the large scanning area to be considered. Accurate localization of impacts could narrow the scanning area and improve testing efficiency. In this paper, the multi-frequency characteristic of impact induced guided waves is analyzed. The time of arrival features at various frequencies are extracted from time domain signal by continuous wavelet transform, cross-correlation and Hilbert transform. Two methods, namely image fusion and artificial neural network, are proposed to accomplish the fusion of multi-frequency features. The results show that the mean localization error for the two fusion methods are 4.39 mm and 3.61 mm respectively, the median error for the two methods are 3.56 mm and 3.13 mm.