Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring
ABSTRACT We provide a condition monitoring system for wind farms, based on normal behaviour modelling using a probabilistic multilayer perceptron with transfer learning via fine‐tuning. The model predicts the output power of the wind turbine under normal behaviour based on features retrieved from su...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-04-01
|
| Series: | Wind Energy |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/we.70012 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850038203943223296 |
|---|---|
| author | Filippo Fiocchi Domniki Ladopoulou Petros Dellaportas |
| author_facet | Filippo Fiocchi Domniki Ladopoulou Petros Dellaportas |
| author_sort | Filippo Fiocchi |
| collection | DOAJ |
| description | ABSTRACT We provide a condition monitoring system for wind farms, based on normal behaviour modelling using a probabilistic multilayer perceptron with transfer learning via fine‐tuning. The model predicts the output power of the wind turbine under normal behaviour based on features retrieved from supervisory control and data acquisition (SCADA) systems. Its advantages are that (i) it can be trained with SCADA data of at least a few years, (ii) it can incorporate all SCADA data of all wind turbines in a wind farm as features, (iii) it assumes that the output power follows a normal density with heteroscedastic variance and (iv) it can predict the output of one wind turbine by borrowing strength from the data of all other wind turbines in a farm. Probabilistic guidelines for condition monitoring are given via a cumulative sum (CUSUM) control chart, which is specifically designed based on a real‐data classification exercise and, hence, is adapted to the needs of a wind farm. We illustrate the performance of our model in a real SCADA data example which provides evidence that it outperforms other probabilistic prediction models. |
| format | Article |
| id | doaj-art-1ea78a43aef549d7980fac8a8f2cfce6 |
| institution | DOAJ |
| issn | 1095-4244 1099-1824 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Wiley |
| record_format | Article |
| series | Wind Energy |
| spelling | doaj-art-1ea78a43aef549d7980fac8a8f2cfce62025-08-20T02:56:39ZengWileyWind Energy1095-42441099-18242025-04-01284n/an/a10.1002/we.70012Probabilistic Multilayer Perceptrons for Wind Farm Condition MonitoringFilippo Fiocchi0Domniki Ladopoulou1Petros Dellaportas2Department of Computer Science University College London London UKDepartment of Statistical Science University College London London UKDepartment of Statistical Science University College London London UKABSTRACT We provide a condition monitoring system for wind farms, based on normal behaviour modelling using a probabilistic multilayer perceptron with transfer learning via fine‐tuning. The model predicts the output power of the wind turbine under normal behaviour based on features retrieved from supervisory control and data acquisition (SCADA) systems. Its advantages are that (i) it can be trained with SCADA data of at least a few years, (ii) it can incorporate all SCADA data of all wind turbines in a wind farm as features, (iii) it assumes that the output power follows a normal density with heteroscedastic variance and (iv) it can predict the output of one wind turbine by borrowing strength from the data of all other wind turbines in a farm. Probabilistic guidelines for condition monitoring are given via a cumulative sum (CUSUM) control chart, which is specifically designed based on a real‐data classification exercise and, hence, is adapted to the needs of a wind farm. We illustrate the performance of our model in a real SCADA data example which provides evidence that it outperforms other probabilistic prediction models.https://doi.org/10.1002/we.70012CUSUM control chartfine‐tuningheteroscedasticitynormal behaviour modellingtransfer learning |
| spellingShingle | Filippo Fiocchi Domniki Ladopoulou Petros Dellaportas Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring Wind Energy CUSUM control chart fine‐tuning heteroscedasticity normal behaviour modelling transfer learning |
| title | Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring |
| title_full | Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring |
| title_fullStr | Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring |
| title_full_unstemmed | Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring |
| title_short | Probabilistic Multilayer Perceptrons for Wind Farm Condition Monitoring |
| title_sort | probabilistic multilayer perceptrons for wind farm condition monitoring |
| topic | CUSUM control chart fine‐tuning heteroscedasticity normal behaviour modelling transfer learning |
| url | https://doi.org/10.1002/we.70012 |
| work_keys_str_mv | AT filippofiocchi probabilisticmultilayerperceptronsforwindfarmconditionmonitoring AT domnikiladopoulou probabilisticmultilayerperceptronsforwindfarmconditionmonitoring AT petrosdellaportas probabilisticmultilayerperceptronsforwindfarmconditionmonitoring |