Liu, Z.-H., Li, L.-W., Wei, H.-L. orcid.org/0000-0002-4704-7346 et al. (3 more authors) (2025) Periodic-enhanced informer model for short-term wind power forecasting using SCADA data. IEEE Transactions on Sustainable Energy. ISSN 1949-3029
Abstract
Supervisory Control and Data Acquisition (SCADA) systems can collect abundant information about wind farm operation and environment. To better make use of SCADA data, a periodic-enhanced informer model for short-term wind power forecasting using scada data is proposed. Firstly, to effectively filter out noise in SCADA data, a v-p curve-based method is adopted by incorporating a quartile approach to remove sparse outliers; a density-based spatial clustering of applications with noise (DBSCAN) algorithm is then employed to eliminate stacked outliers from the power curve. Secondly, a multi-feature input set selection method based on Maximization Information Coefficient is introduced to make better use of the SCADA system data by reducing the number of features. Thirdly, a Temporal Convolutional Network (TCN) is designed to extract the scalar projection of the input set, followed by fusing the local time stamp and global time stamp to build the periodic information enhanced prediction model embedding layer. Subsequently, the enhanced input set is fed into an informer model to predict future wind power. Finally, considering the multiple temporal scales structure characteristics present in the dataset, a multi-scale deep fusion module is established in the informer model to deeply integrate the features of different scales. It can simultaneously avoid the resource waste and overfitting problems caused by increasing the network depth. The experimental results, which are obtained from several deep learning methods on real SCADA data, demonstrate that the suggested approach has good predictive capability.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2025 The Author(s). Except as otherwise noted, this author-accepted version of a journal article published in IEEE Transactions on Sustainable Energy is made available via the University of Sheffield Research Publications and Copyright Policy under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution and reproduction in any medium, provided the original work is properly cited. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ |
Keywords: | SCADA; wind power forecasting; TCN; informer; maximum information coefficient |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > School of Electrical and Electronic Engineering |
Funding Information: | Funder Grant number ROYAL SOCIETY IEC\NSFC\223266 |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 08 Apr 2025 12:38 |
Last Modified: | 10 Apr 2025 08:41 |
Status: | Published online |
Publisher: | Institute of Electrical and Electronics Engineers |
Refereed: | Yes |
Identification Number: | 10.1109/TSTE.2025.3558726 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:225276 |
Download
Filename: 2025 TSTE-00890-2024 Final Accepted Manuscript.pdf
Licence: CC-BY 4.0