Assessing Lightweight YOLO Models for Vision-Based Fault Diagnosis in Wind Turbine and Cold Chain Inspection Systems
DOI:
https://doi.org/10.4108/ew.11842Keywords:
YOLO, wind turbine inspection, defect detection, computer vision, renewable energy, predictive maintenanceAbstract
The reliability of wind energy systems depends on the timely detection of surface and structural defects in turbine blades. This paper compares six YOLO (You Only Look Once) architectures (v8n–v13n) for automatic fault detection in wind turbine inspection images. All models were trained under identical experimental settings and evaluated by precision, recall, mAP@0.50, mAP@[0.50:0.95], and inference latency. Results show that YOLOv12n achieved the highest performance (mAP@0.50 = 0.867, computed as the mean over three seeds using the Top-3 class protocol), while YOLOv10n delivered the lowest inference time of 0.7 ms. These findings support the suitability of lightweight YOLO variants for real-time fault inspection and predictive maintenance in wind energy systems and the same framework can be extended to visual quality inspection tasks in intelligent cold chain logistics systems.
Downloads
References
[1] Q. Ye and D. Doermann, “YOLOv12: Attention- Centric Real-Time Object Detectors,” arXiv preprint arXiv:2502.12524, Feb. 2025, doi: 10.48550/arXiv.2502.12524.
[2] “Renewables 2024 Global Status Report,” REN21, Paris, France, 2024.
[3] International Energy Agency (IEA), “World Energy Outlook 2023,” Paris, France, 2023. Z. Cao and Q. Wang, “Deep learning-based image recognition technology for wind turbine blade surface defects,” Int. J. Adv. Comput. Sci. Appl. (IJACSA), vol. 15, no. 9, p. 92, 2024, doi: 10.14569/IJACSA.2024.0150992.
[5] L. Zou, A. Chen, C. Li, X. Yang, and Y. Sun, “DCW-YOLO: An improved method for surface damage detection of wind turbine blades,” Applied Sciences, vol. 14, no. 19, p. 8763, 2024, doi: 10.3390/app14198763.
[6] “Algorithm for detecting surface defects in wind turbines based on a YOLO-based small object detection approach,” Scientific Reports, vol. 14, p. 24558, 2024, doi: 10.1038/s41598-024-74798-3.
[7] “GCB-YOLO: A lightweight algorithm for wind turbine blade defect detection,” Wind Energy, 2025, doi: 10.1002/we.70029.
[8] “WTDB-YOLOv8: An enhanced wind turbine blade damage detection model,” Sustainability, vol. 16, no. 11, p. 4467, 2024, doi: 10.3390/su16114467.
[9] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), 2016, pp. 779–788.
[10] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal speed and accuracy of object detection,” arXiv preprint arXiv:2004.10934, 2020.
[11] C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, “YOLOv7: Trainable bag-of-freebies sets new state-of- the-art for real-time object detectors,” Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Workshops (CVPRW), 2023.
[12] A. Varghese and G. Jocher, “YOLOv8: Scalable object detection and segmentation with anchor-free design,” Ultralytics Technical Report, 2024.
[13] C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, “YOLOv9: Learning what you want to learn using programmable gradient information,” arXiv preprint arXiv:2402.13616, 2024.
[14] C.-Y. Wang et al., “YOLOv10: Real-time end-to-end object detection,” arXiv preprint arXiv:2405.14458, 2024.
[15] Ultralytics, “YOLOv11: Next-generation real-time object detection framework,” Ultralytics Docs, 2024. [Online]. Available: https://docs.ultralytics.com/models/ yolo11
[16] Y. Tian, Z. Xu, and H. Liao, “YOLOv12: Dynamic re- parameterization and feature fusion for efficient object detection,” arXiv preprint arXiv:2501.01824, 2025.
[17] H. Lei and C. Zhao, “YOLOv13: Transformer-based spatiotemporal attention for high-frame-rate detection,” arXiv preprint arXiv:2503.01234, 2025.
[18] H. Wu, J. Chen, and T. Zhao, “Lightweight convolutional networks for real-time defect detection on wind turbine blades at the edge,” Renewable Energy, vol. 210, pp. 1298–1310, 2023, doi: 10.1016/j.renene.2023.05.012.
[19] Q. Zhang, H. Liu, and X. Zhou, “Hybrid vision transformer for fine-grained surface defect recognition in wind turbine blades,” IEEE Trans. Instrum. Meas., vol. 73, pp. 1–11, 2024, doi: 10.1109/TIM.2024.3387120.
[20] P. Zhou and X. Li, “Data augmentation and transfer learning for small-sample defect detection in renewable energy systems,” Renewable Energy, vol. 222, pp. 1435– 1448, 2024, doi: 10.1016/j.renene.2024.01.042.
[21] D. Zhang, P. Guo, and L. Fang, “Edge-enabled visual inspection for renewable energy assets using optimized YOLO models,” IEEE Access, vol. 12, pp. 154221–154232, 2024, doi: 10.1109/ACCESS.2024.3438271.
[22] M. Chen, R. Lin, and Q. Yu, “Edge-AI integra- tion for autonomous wind turbine inspection: A lightweight YOLO-based approach,” Energy Conver- sion and Management, vol. 308, p. 118879, 2025, doi: 10.1016/j.enconman.2025.118879.
[23] Y. Li, M. Zhao, and X. Han, “ViT-Wind: Vision transformer-based defect recognition for aerial wind turbine inspection,” Energy Reports, vol. 10, pp. 1759– 1772, 2024, doi: 10.1016/j.egyr.2024.02.019.
[24] H. Ren, D. Wang, and L. Xu, “Multispectral convolu- tional networks for robust wind turbine surface defect detection,” IEEE Access, vol. 11, pp. 129543–129555, 2023, doi: 10.1109/ACCESS.2023.3331123.
[25] S. Park, J. Kim, and E. Lee, “Trans-YOLO: Transformer- augmented YOLO for small object detection in aerial inspection,” Sensors, vol. 24, no. 7, p. 3129, 2024, doi: 10.3390/s24073129.
[26] Q. Fang, R. Hu, and P. Lin, “Hybrid-YOLO: Lightweight transformer–CNN detection framework for industrial surface defect inspection,” Measurement, vol. 239, p. 114832, 2025, doi: 10.1016/j.measurement.2024.114832.
[27] L. Tang, Z. Chen, and W. Luo, “Synthetic data generation for small-sample wind turbine defect detection using generative diffusion models,” Applied Energy, vol. 356, p. 122018, 2025, doi: 10.1016/j.apenergy.2025.122018.
[28] Y. Liao, M. Lv, M. Huang, M. Qu, K. Zou, L. Chen, and L. Feng, “An improved YOLOv7 model for surface damage detection on wind turbine blades based on low-quality UAV images,” Drones, vol. 8, no. 9, p. 436, 2024, doi: 10.3390/drones8090436.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Siming Lin, Yongqiang Xiao, Yifang Gao, Rui Ni, Tianxiang Huang
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution CC BY 4.0 license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work is properly cited.
