dc.contributor.author | Keprate, Arvind | |
dc.contributor.author | Sheikhi, Saeid | |
dc.contributor.author | Siddiqui, Muhammad Salman | |
dc.date.accessioned | 2024-02-02T13:29:31Z | |
dc.date.available | 2024-02-02T13:29:31Z | |
dc.date.created | 2024-02-01T13:38:34Z | |
dc.date.issued | 2023 | |
dc.identifier.isbn | 9798350323153 | |
dc.identifier.isbn | 979-8-3503-2316-0 | |
dc.identifier.uri | https://hdl.handle.net/11250/3115336 | |
dc.description.abstract | Offshore wind turbines (OWTs) play a crucial role in renewable energy generation, but their remote and harsh environments make them prone to various anomalies that can significantly affect their performance and reliability. This article compares deep learning-based image processing techniques for unsupervised anomaly detection in OWTs. Initially, an investigation into three signal-to-image encoding algorithms namely Gramian Angular Summation Field, and Gramian Angular Difference Field, and Markov Transition Field to transform time series data into image-like representations. The study demonstrates that the choice of encoding technique significantly influences the outcomes when employed in deep learning architectures. The evaluation uses a generator-bearing dataset of an offshore wind turbine located in Africa. The results reveal that certain encodings exhibit a competitive edge and should be considered when applying deep learning frameworks for anomaly detection. In conclusion, this research underscores the potential of deep learning and image-like representations in effectively identifying anomalies in time series data. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | en_US |
dc.relation.ispartof | IEEM 2023 IEEE International Conference on Industrial Engineering and Engineering Management Singapore 18 - 21 December 2023 | |
dc.relation.ispartofseries | IEEE International Conference on Industrial Engineering and Engineering Management; | |
dc.title | Comparing Deep Learning Based Image Processing Techniques for Unsupervised Anomaly Detection in Offshore Wind Turbines | en_US |
dc.type | Chapter | en_US |
dc.type | Peer reviewed | en_US |
dc.type | Conference object | en_US |
dc.description.version | acceptedVersion | en_US |
cristin.ispublished | true | |
cristin.fulltext | preprint | |
cristin.qualitycode | 1 | |
dc.identifier.doi | https://doi.org/10.1109/IEEM58616.2023.10406361 | |
dc.identifier.cristin | 2241930 | |