Ice Detection on Rotor Blades of Wind Turbines using RGB Images and Convolutional Neural Networks
| Autor: | Klaus-Dieter Thoben, Michael Freitag, Markus Kreutz, Abderrahim Ait Alla, Anatoli Eisenstadt |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
0209 industrial biotechnology
Wind power Ice formation Nacelle Rotor (electric) business.industry Computer science 02 engineering and technology 010501 environmental sciences 01 natural sciences Convolutional neural network law.invention 020901 industrial engineering & automation law General Earth and Planetary Sciences RGB color model business 0105 earth and related environmental sciences General Environmental Science Marine engineering |
| Zdroj: | Procedia CIRP. 93:1292-1297 |
| ISSN: | 2212-8271 |
| DOI: | 10.1016/j.procir.2020.04.107 |
| Popis: | Ice formed on wind turbines not only causes economic damage but also poses a risk to nearby humans. Countermeasuring ice formation requires reliable detection systems. Existing ice detection solutions have improved with time, but are still not sufficiently accurate. This paper investigates the use of different convolutional neural networks (CNNs) and RGB images taken from cameras installed on the nacelle of wind turbines to detect ice on the rotor blades. In our research, the VGG19 model achieved the best performance with an accuracy and an f1-score of (96 ± 2) %. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|