Vibration Based - Crack Detection in Simplified Wind Turbine Blades Using Artificial Neural Networks

Abstract

Wind turbine blades are complicated components for inspection by non-destructive techniques because they are multi-layered, have variable thickness and are made of anisotropic materials. This paper proposes the use of Artificial Neural Networks (ANN) for the detection of crack location and crack depth in wind turbine blades. Wind turbine blade is approximated by a laminated composite, cantilever tapered beam with a transverse open surface crack. The natural frequencies which are influenced by crack specifications are obtained by a Finite Element Method (FEM) via ANSYS software. Experimental setup has been developed to validate the results obtained from the finite element software ANSYS. The numerical data obtained from (FEM) are then used to train a feed-forward back propagation neural network using Matlab environment. The input parameters to the neural network are the first three relative natural frequencies, while the output parameters are the relative crack depth and relative crack location. Simulations are carried out to test the performance and the accuracy of the trained network by comparing the results for the crack depth and crack location obtained from (ANN) with those obtained from (FEM). The simulation results show that the proposed Artificial Neural Network can precisely detect the crack location and crack depth.

Keywords

• crack detection
• wind turbine blade
• composite beam
• neural network