Acoustic emission data based deep learning approach for classification and detection of damage-sources in a composite panel

Structural health monitoring for lightweight complex composite structures is being investigated in this paper with a data-driven deep learning approach to facilitate automated of the map transformed signal features damage classes. Towards this, series acoustic emission (AE) based laboratory experiments have been carried out on sample using piezoelectric AE sensor network. The registered time-domain signals from assigned networks panel are processed continuous wavelet transform extract time-frequency scalograms . A convolutional neural network architecture proposed automatically discrete scalogram images and use them classify damage-source regions panel. deep-learning has shown an effective potential high training, validation test accuracy unseen datasets as well entirely new neighboring datasets. Further, trained, validated tested only peak-signal data extracted raw data. application significant improvement classification performance accuracy.