Abstract
Abstract: Breaking the curse of small datasets in machine learning is but one of the major challenges that cause several real-life prediction problems. In offshore wind application, for instance, this issue presents when monitoring an asset in an attempt to reduce its infant mortality failures. Another challenge could emerge when reducing the number of sensors installed in order to limit the investment in monitoring systems. To tackle these issues, the aim of this article is to investigate the impact of small data-set on conventional machine learning methods, and to outline the improvement achievable by the implementation of transfer learning approach. It provides a solution to mitigate this issue by applying a hard parameter multi-task learning approach to the supervisory control and data acquisition data from an operational wind turbine, allowing for smaller datasets to efficiently predict the status of the gearbox's vibration data. Two experiments are carried out in this paper. The first is to envisage the possibility of using hard parameter transfer on the operational data from two wind turbines. The second is to compare the results of this model to the findings from a conventional deep neural network model trained on the data from a single turbine.
Original language | English |
---|---|
Article number | 032091 |
Number of pages | 10 |
Journal | Journal of Physics: Conference Series |
Volume | 2265 |
Issue number | 3 |
DOIs | |
Publication status | Published - 1 May 2022 |
Event | The Science of Making Torque from Wind - TU Delft, Delft, Netherlands Duration: 1 Jun 2022 → 3 Jun 2022 Conference number: 9th https://www.torque2022.eu/ |
Keywords
- classification
- deep neural network (DNN)
- hard parameter transfer
- multi-task learning
- condition monitoring
- offshore wind turbine
- supervisory control and data acquisition (SCADA)
- condition monitoring systems (CMS)
- machine learning (ML)
- regression