Prediction of fatigue damage growth in notched composite laminates using an artificial neural network

Models to predict the split growth in notched AS4/3501-6 graphite/epoxy quasi-isotropic laminates under tension-dominated fatigue are presented. First, a power law model and an artificial neural network (ANN) model are developed to describe the split growth under constant-amplitude fatigue. They are then applied in conjunction with a linear damage growth model to predict the split growth under spectrum fatigue. The ANN model is found to work better than the power law model as a predictive tool for split growth.