Day-Ahead Wind Power Forecasting Using a Two-Stage Hybrid Modeling Approach Based on SCADA and Meteorological Information, and Evaluating the Impact of Input-Data Dependency on Forecasting Accuracy

Dehua Zheng, Min Shi, Yifeng Wang, Abinet Tesfaye Eseye and Jianhua Zhang
Additional contact information
Dehua Zheng: Microgrid Platform R&D Center, Goldwind Science and Etechwin Electric Co., Ltd. BDA, Beijing 100176, China
Min Shi: State Grid Hebei Electric Power Company, Shijiazhuang 050022, China
Yifeng Wang: State Grid Hebei Electric Power Company, Shijiazhuang 050022, China
Abinet Tesfaye Eseye: Microgrid Platform R&D Center, Goldwind Science and Etechwin Electric Co., Ltd. BDA, Beijing 100176, China
Jianhua Zhang: School of Electrical and Electronic Engineering, North China Electric Power University, Changping District, Beijing 102206, China

Energies, 2017, vol. 10, issue 12, 1-23

Abstract: The power generated by wind generators is usually associated with uncertainties, due to the intermittency of wind speed and other weather variables. This creates a big challenge for transmission system operators (TSOs) and distribution system operators (DSOs) in terms of connecting, controlling and managing power networks with high-penetration wind energy. Hence, in these power networks, accurate wind power forecasts are essential for their reliable and efficient operation. They support TSOs and DSOs in enhancing the control and management of the power network. In this paper, a novel two-stage hybrid approach based on the combination of the Hilbert-Huang transform (HHT), genetic algorithm (GA) and artificial neural network (ANN) is proposed for day-ahead wind power forecasting. The approach is composed of two stages. The first stage utilizes numerical weather prediction (NWP) meteorological information to predict wind speed at the exact site of the wind farm. The second stage maps actual wind speed vs. power characteristics recorded by SCADA. Then, the wind speed forecast in the first stage for the future day is fed to the second stage to predict the future day’s wind power. Comparative selection of input-data parameter sets for the forecasting model and impact analysis of input-data dependency on forecasting accuracy have also been studied. The proposed approach achieves significant forecasting accuracy improvement compared with three other artificial intelligence-based forecasting approaches and a benchmark model using the smart persistence method.

Keywords: artificial neural network; forecasting; genetic algorithm; Hilbert-Huang transform; NWP (numerical weather prediction); SCADA (supervisory control and data acquisition); wind power (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2017
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (10)

Downloads: (external link)
https://www.mdpi.com/1996-1073/10/12/1988/pdf (application/pdf)
https://www.mdpi.com/1996-1073/10/12/1988/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:10:y:2017:i:12:p:1988-:d:121516

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().