A comparative study of CNN and vision transformer architectures for fault detection in small wind turbine blades

Öz

Automated fault detection in wind turbine blades is critical for ensuring the reliability and operational efficiency of wind energy systems. This study presents a systematic comparative analysis of five state-of-the-art deep learning architectures for binary fault classification (healthy versus faulty) on the CAI-SWTB dataset, comprising 6,000 RGB images of small wind turbine blades. The evaluated architectures span three distinct design paradigms: a classical Convolutional Neural Network (CNN) (ResNet-50), a modern CNN (EfficientNetV2-S), and three Vision Transformer (ViT)-based models (ViT-B/16, Data-efficient Image Transformer (DeiT)-S/16, and Swin-Tiny). All models were trained using a two-stage transfer learning protocol with ImageNet-pretrained weights, employing the AdamW optimizer and a cosine annealing learning rate schedule. EfficientNetV2-S achieved the highest classification performance with 99.75% accuracy, followed by ResNet-50 at 99.42%. Among the transformer-based models, Swin-Tiny outperformed both ViT-B/16 (65.33%) and DeiT-S/16 (82.75%), achieving 88.25% accuracy. Grad-CAM analysis confirmed that the best-performing models correctly localize structural defect regions in blade images, supporting their interpretability and suitability for real-world inspection applications.

Anahtar Kelimeler

• Wind turbine blade
• Fault detection
• Convolutional neural network
• Vision transformer

Kaynakça

1. [1] Du Y, Zhou S, Jing X, Peng Y, Wu H, Kwok N. Damage detection techniques for wind turbine blades: A review. Mechanical Systems and Signal Processing 2020; 141: 106445.
2. [2] Kaewniam P, Cao M, Alkayem NF, Li D, Manoach E. Recent advances in damage detection of wind turbine blades: A state-of-the-art review. Renewable and Sustainable Energy Reviews 2022; 167: 112723.
3. [3] Mishnaevsky Jr L, Thomsen K. Costs of repair of wind turbine blades: Influence of technology aspects. Wind Energy 2020; 23: 2247-2255.
4. [4] Shihavuddin A, Chen X, Fedorov V, Nymark Christensen A, Andre Brogaard Riis N, Branner K, Bjorholm Dahl A, Reinhold Paulsen R. Wind turbine surface damage detection by deep learning aided drone inspection analysis. Energies 2019; 12: 676.
5. [5] Ye X, Wang L, Huang C, Luo X. Wind turbine blade defect detection with a semi-supervised deep learning framework. Engineering Applications of Artificial Intelligence 2024; 136: 108908.
6. [6] Altice B, Nazario E, Davis M, Shekaramiz M, Moon TK, Masoum MA. Anomaly detection on small wind turbine blades using deep learning algorithms. Energies 2024; 17: 982.
7. [7] Mehadjbia A, Slaoui-Hasnaoui F. Wind Turbine surface defects detection Based-on NasNetMobile architecture. 2024 International Conference on Electrical, Computer and Energy Technologies (ICECET), IEEE 2024; 1-6.
8. [8] Gonçalves L, Rego CB, Bezerra PHA. Comparing performances of deep learning models to classify and detect defects in wind turbine blades. Congresso Brasileiro de Inteligência Computacional 2025; 7.

9. [9] Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, Dehghani M, Minderer M, Heigold G, Gelly S. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 2020.
10. [10] Touvron H, Cord M, Douze M, Massa F, Sablayrolles A, Jégou H. Training data-efficient image transformers & distillation through attention. International conference on machine learning, PMLR 2021: 10347-10357.
11. [11] Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, Lin S, Guo B. Swin transformer: Hierarchical vision transformer using shifted windows. Proceedings of the IEEE/CVF international conference on computer vision 2021: 10012-10022.
12. [12] Tan M, Le Q. EfficientNetV2: Smaller models and faster training. International conference on machine learning, PMLR 2021: 10096-10106.
13. [13] Zhang C, Yang T, Yang J. Image recognition of wind turbine blade defects using attention-based MobileNetv1-YOLOv4 and transfer learning. Sensors 2022; 22: 6009.
14. [14] Guo J, Liu C, Cao J, Jiang D. Damage identification of wind turbine blades with deep convolutional neural networks. Renewable Energy 2021; 174: 122-133.
15. [15] Davis M, Nazario Dejesus E, Shekaramiz M, Zander J, Memari M. Identification and Localization of Wind Turbine Blade Faults Using Deep Learning. Applied Sciences 2024; 14: 6319.
16. [16] Wang Q, Zhang H, Wang X, Ni Y. Wind Turbine Airfoil Noise Prediction Method Based on Generalized Airfoil Database and Residual Neural Network. Applied Sciences 2025; 15: 5123.
17. [17] Dwivedi D, Babu KVSM, Yemula PK, Chakraborty P, Pal M. Identification of surface defects on solar PV panels and wind turbine blades using attention based deep learning model. Engineering Applications of Artificial Intelligence 2024; 131: 107836.
18. [18] Jeevan P, Sethi A. Which backbone to use: a resource-efficient domain specific comparison for computer vision. arXiv preprint arXiv:2406.05612 2024.
19. [19] Takahashi S, Sakaguchi Y, Kouno N, Takasawa K, Ishizu K, Akagi Y, Aoyama R, Teraya N, Bolatkan A, Shinkai N. Comparison of vision transformers and convolutional neural networks in medical image analysis: a systematic review. Journal of Medical Systems 2024; 48: 84.
20. [20] He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition 2016: 770-778.
21. [21] Loshchilov I, Hutter F. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101 2017.
22. [22] Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE international conference on computer vision 2017: 618-626.