Abstract
Undetected micro-cracks in wind turbine blades lead to failure and costly repairs. Effective remedies depend on proactive measures taken to detect such cracks at early stages. Defects and cracks cause certain shifts in resonant peaks amplitude and frequency (spectral shifts) of the blade depending on the location and dimensions of the crack or damage. Precise location of the crack is difficult to ascertain due to its small dimensions and random causes. This study developed a special Kriging model for predicting crack location using spectral shifts in the blade’s resonant peaks. Records of blade’s spectral shifts collected for known crack locations were used to construct the Kriging model. The model predicts crack location by correlating spectral shifts of cracked blade with spectral shifts recorded from blades with known crack locations. Experimental validation demonstrated the ability of the method to detect micro crack locations with an accuracy of 95%. Experimental results are presented and discussed.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call