This is a work of Data Science Lab of ETH, the proejct was primarily conducted by Zhengxu Li and Lucien under the supervision of Dr.Imad Abdallah and PD. Dr. Alexander Ilic.
This paper investigates the use of machine learning models for under- standing the severity of damage in wind turbine blades. Various feature extraction methods are explored and a range of classifiers are employed to optimize accuracy and recall. This research finds that the combination of Minirocket with the CatBosst classifier achieves the best performance. This research demonstrates the efficacy of these systems in improving the efficiency and reducing the cost of wind turbine maintenance.