Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy
The quality of wind data from the numerical weather prediction significantly influences the accuracy of wind power forecasting systems for wind parks. Therefore, an in-depth investigation of these wind data themselves is essential to improve wind power generation efficiency and maintain grid reliabi...
Published in: | Energy Reports |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Elsevier
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10037/26391 https://doi.org/10.1016/j.egyr.2022.08.105 |
id |
ftunivtroemsoe:oai:munin.uit.no:10037/26391 |
---|---|
record_format |
openpolar |
spelling |
ftunivtroemsoe:oai:munin.uit.no:10037/26391 2023-05-15T15:04:16+02:00 Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy Chen, Hao Zhang, Qixia Birkelund, Yngve 2022-08-22 https://hdl.handle.net/10037/26391 https://doi.org/10.1016/j.egyr.2022.08.105 eng eng Elsevier Energy Reports Chen, Zhang, Birkelund. Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy. Energy Reports. 2022 FRIDAID 2045483 doi:10.1016/j.egyr.2022.08.105 2352-4847 https://hdl.handle.net/10037/26391 openAccess Copyright 2022 The Author(s) Journal article Tidsskriftartikkel Peer reviewed publishedVersion 2022 ftunivtroemsoe https://doi.org/10.1016/j.egyr.2022.08.105 2022-08-24T23:00:01Z The quality of wind data from the numerical weather prediction significantly influences the accuracy of wind power forecasting systems for wind parks. Therefore, an in-depth investigation of these wind data themselves is essential to improve wind power generation efficiency and maintain grid reliability. This paper proposes a novel framework based on machine learning for concurrently analyzing and forecasting predictive errors, called residuals, of wind speed and direction from a numerical weather prediction model versus measurements over a while. The performance of the framework is testified by a wind farm inside the Arctic. It is demonstrated that the residuals still contain significant meteorological information and can be effectively predicted with machine learning and the linear autoregression works well for multi-timesteps predictions of overall, East-West, East–West, and North-South North–South wind speeds residuals by comparing the four forecast learning algorithms’ performance. The predictions may be applied to correct the NWP wind model, making quality feedback improvements for inputs for wind power forecasting systems. Article in Journal/Newspaper Arctic University of Tromsø: Munin Open Research Archive Arctic Energy Reports 8 661 668 |
institution |
Open Polar |
collection |
University of Tromsø: Munin Open Research Archive |
op_collection_id |
ftunivtroemsoe |
language |
English |
description |
The quality of wind data from the numerical weather prediction significantly influences the accuracy of wind power forecasting systems for wind parks. Therefore, an in-depth investigation of these wind data themselves is essential to improve wind power generation efficiency and maintain grid reliability. This paper proposes a novel framework based on machine learning for concurrently analyzing and forecasting predictive errors, called residuals, of wind speed and direction from a numerical weather prediction model versus measurements over a while. The performance of the framework is testified by a wind farm inside the Arctic. It is demonstrated that the residuals still contain significant meteorological information and can be effectively predicted with machine learning and the linear autoregression works well for multi-timesteps predictions of overall, East-West, East–West, and North-South North–South wind speeds residuals by comparing the four forecast learning algorithms’ performance. The predictions may be applied to correct the NWP wind model, making quality feedback improvements for inputs for wind power forecasting systems. |
format |
Article in Journal/Newspaper |
author |
Chen, Hao Zhang, Qixia Birkelund, Yngve |
spellingShingle |
Chen, Hao Zhang, Qixia Birkelund, Yngve Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
author_facet |
Chen, Hao Zhang, Qixia Birkelund, Yngve |
author_sort |
Chen, Hao |
title |
Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
title_short |
Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
title_full |
Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
title_fullStr |
Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
title_full_unstemmed |
Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
title_sort |
machine learning forecasts of scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
publisher |
Elsevier |
publishDate |
2022 |
url |
https://hdl.handle.net/10037/26391 https://doi.org/10.1016/j.egyr.2022.08.105 |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic |
genre_facet |
Arctic |
op_relation |
Energy Reports Chen, Zhang, Birkelund. Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy. Energy Reports. 2022 FRIDAID 2045483 doi:10.1016/j.egyr.2022.08.105 2352-4847 https://hdl.handle.net/10037/26391 |
op_rights |
openAccess Copyright 2022 The Author(s) |
op_doi |
https://doi.org/10.1016/j.egyr.2022.08.105 |
container_title |
Energy Reports |
container_volume |
8 |
container_start_page |
661 |
op_container_end_page |
668 |
_version_ |
1766336066977005568 |