The simulation for damage identification of wing model basis on PNN

Based on the finite element wing model of a certain UAV, neural network approach is used to identify the location and degree of damage. First of all, the theory, characteristics and structure of probabilistic neural network (PNN) is introduced. Then, the one-dimensional and two-dimensional simplified finite element damage models of the wing are established. By the extraction and processing of their natural frequencies and vibration modes from damaged wing structure with different damage position and different damage degree, the trained PNN have the ability of damage identification. Simulation shows that, the model based on PNN can effectively identify the location and degree of damage.