Dynamic Study and Detection of Edge Crack in Composite Laminates Using Vibration Parameters

Composite materials are used as structural components in many disciplines. The structures may become prone to damage during the service conditions. The damage types are surface crack, edge crack, and delamination. The damage may lead to abrupt failure of the big composite structures. To prevent the abrupt failure, the damage should be detected at the earliest by monitoring the damage indicators. The change in natural frequency of the structure is considered as one of the damage indicators. The cantilever beam is considered here to understand the natural frequency variation with crack location and crack length. The natural frequency decreases with increasing edge crack length and the natural frequency reduction is more toward the fixed end compared to the free end of the beam. The crack location and length of an edge crack of size 1 mm have also been predicted by using the feed forward neural network. The inputs to the neural network were frequency change rates and the outputs were crack location and length. Neural network of 3-12-2 architecture is able to predict the crack location and length with more than 94% accuracy. The accuracy of the location of the edge crack prediction is more when compared to the crack length prediction.