Wind turbine blade fault detection and identification based on improved YOLOv5s

Xianwen Da; Li Lu

doi:10.1117/12.3029141

6 May 2024 Wind turbine blade fault detection and identification based on improved YOLOv5s

Xianwen Da, Li Lu

Proceedings Volume 13107, Fourth International Conference on Sensors and Information Technology (ICSI 2024); 131073A (2024) https://doi.org/10.1117/12.3029141
Event: Fourth International Conference on Sensors and Information Technology (ICSI 2024), 2024, Xiamen, China

Abstract

Failure of wind turbine blades is a major problem affecting the sustainable and healthy development of wind power generation, and also hides huge safety hazards and environmental problems. In order to detect the faults of wind turbine blades, this paper proposes a wind turbine blade fault identification and detection method based on the improved YOLOv5s algorithm. Firstly, a variable convolutional network is used to replace the ordinary convolutional network in the feature extraction network of the original YOLOv5s, which makes the model lightweight and maintains good detection accuracy; secondly, the Wise Iou loss function is used to solve the imbalance problem of the defect categories in the dataset and to make the target detection model converge; lastly, tests are carried out in the dataset of the wind turbine blades and the experimental results are compared with those of the other datasets. The experimental results show that the algorithm is effective in detecting defects. The experimental results show that the detection accuracy of the algorithm reaches 93.6%, which verifies the effectiveness and accuracy of the improved YOLOv5-based algorithm.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Xianwen Da and Li Lu "Wind turbine blade fault detection and identification based on improved YOLOv5s", Proc. SPIE 13107, Fourth International Conference on Sensors and Information Technology (ICSI 2024), 131073A (6 May 2024); https://doi.org/10.1117/12.3029141

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available