Damage identification of wind turbine blades with deep convolutional neural networks.

Online early detection of surface damages on blades is critical for the safety of wind turbines, which could avoid catastrophic failures, minimize downtime, and enhance the reliability of the system. Monitoring the health status of blades is attracting more and more attention including on-site cameras and mobile cameras by drones and crawling robots. To deploy fast and efficient damage detection methods from image data, this work presents a hierarchical identification framework for wind turbine blades, which consists of a Haar-AdaBoost step for region proposal and a convolutional neural network (CNN) classifier for damage detection and fault diagnosis. Case studies are carried out on real data set collected from an eastern China wind farm. Results show that (i) the proposed framework can detect and identify the blade damages and outperforms other schemes include SVM and VGG16 models, (ii) sensitive analysis is conducted to validate the robustness of proposed method under limited data conditions, (iii) the proposed scheme is faster than one-step CNN method that directly classifying raw data. • Hierarchical damage identification framework using images for wind turbine blades. • Classifying defects by Deep CNNs using proposed regions by AdaBoost cascade classifier. • Verified efficient and accurate workflow by balancing false alarm and missed detection. • End-to-end scheme with no need of handcrafted features and applicable to edge devices. [ABSTRACT FROM AUTHOR]