Damage Size Classification of Natural Fibre Reinforced Composites Using Neural Network

Abstract

Damage classification is considered as an important feature in pattern recognition, which led to providing significant information. This research work explores damage size classification for several impact events in natural fibre reinforced composites, which is based on the information provided by the ten piezoceramics (PZT) sensors. An Impact event produced strain waves which several data features were obtained through the response captured. An effective impact damage classification procedure is established using a multilayer perceptron neural network approach. The system was trained to predict the damage size based on the actual experimental data. The data features were mapped into five output class labels, presented as a target confusion matrix. The classification results revealed that the damage sizes were successfully mapped according to its respective class, with the peak to peak feature gives the highest classification rate at 98.4%.