Prediction of impact damage tolerance of drop impacted WGFRP composite by artificial neural network using acoustic emission parameters

Monitoring of drop impact damage is necessary because it produces invisible damage in composite materials without any visible mark on the surface. Monitoring of drop impact damage was carried out on Woven Glass Fibre Reinforced Polymer (WGFRP) composite laminate through Acoustic Emission (AE) technique. The significant AE parameters like signal strength, root means square value, counts and counts to peak were determined for drop impact damage. Impact damage tolerance was predicted using Artificial Neural Network (ANN) trained with AE parameters as input and impact damage tolerance as output. The predicated impact damage tolerance was with average error tolerance of 3.35%. The proposed network finds very good agreement for prediction of impact damage tolerance of impact damaged WGFRP composite laminate in real time application. (C) 2014 Elsevier Ltd. All rights reserved.