Lamb Wave and GA-BP Neural Network Based Damage Identification for Wind Turbine Blade

In order to improve the safety hazard and reduce maintenance cost of wind turbine blades, the damage identification method have been studied by using GA-BP neural network in which BP network is optimized by genetic algorithm. On the basis of the propagation of Lamb waves, composite plate was adopted as it is widely used in blade. And different degrees of damages were artificially made. A specific piezoelectric wafer sensors array was designed for the excitation and collection of Lamb waves propagated in the blades. The Lamb wave responses were collected and characteristic parameters were extracted at different damages. The GA-BP neural network was trained by means of samples so that it might assess the extent of new damage. The experimental results shown that the damage degree can clearly be evaluated.