Machine learning is revolutionizing quality assurance for wind turbines, offering a cutting-edge solution to prevent costly defects. A new approach developed by EPFL and University of Glasgow researchers combines radar technology and artificial intelligence to detect anomalies beneath the surface of wind turbine blades.
This groundbreaking method addresses the limitations of traditional surface inspections, which may overlook critical internal flaws in composite structures. With the use of a patented Frequency Modulated Continuous Wave radar, the research teams have successfully identified features and precursors to potential failures in complex composite samples.
Olga Fink, head of the Intelligent Maintenance and Operations Systems Laboratory at EPFL, emphasizes the significance of this innovation in the face of evolving wind turbine designs. Manufacturers are constructing larger turbines with intricate designs, raising the risk of defects during production, she explains. By employing AI-driven systems, engineers can now identify and address these issues proactively.
The University of Glasgow researchers, under the leadership of Prof. David Flynn, have been instrumental in developing prognostics and health management methods for wind turbines. Their expertise in Robotics and Artificial Intelligence (RAI) has contributed to the success of this collaborative project.
Moving forward, the teams plan to further validate their results by collecting additional data and testing the method on operational wind turbines. By incorporating the radar sensor on a robotic arm or drone, they aim to detect manufacturing defects before turbines are deployed, ensuring optimal performance and longevity.
This multidisciplinary approach holds great promise for enhancing quality assurance standards in the wind energy industry. By harnessing the power of machine learning and advanced radar technology, manufacturers can mitigate the risks associated with faulty turbine blades, ultimately improving operational efficiency and reducing maintenance costs.
