Prediction of multi-scale wind power time series based on transformer

Xuefeng Gao; Hao Li; Zhenduo Gao; Xinhong Wang; Yu Shi; Jia Jia

doi:10.1117/12.2673035

25 April 2023 Prediction of multi-scale wind power time series based on transformer

Xuefeng Gao, Hao Li, Zhenduo Gao, Xinhong Wang, Yu Shi, Jia Jia

Proceedings Volume 12598, Eighth International Conference on Energy Materials and Electrical Engineering (ICEMEE 2022); 1259818 (2023) https://doi.org/10.1117/12.2673035
Event: Eighth International Conference on Energy Materials and Electrical Engineering (ICMEE 2022), 2022, Guangzhou, China

Abstract

Due to the volatility of the output of wind farms, the large-scale integration of wind power into the power grid will greatly increase the difficulty of power generation planning in the power system. This brings great challenges to the scheduling and operation of the power system. The accurate prediction of wind power can effectively relieve the pressure of peak regulation and frequency regulation of the power system and improve the wind power consumption capacity. Therefore, this paper improves on the transformer and proposes a new multi-scale wind power time series forecasting method. First of all, this paper adopts the network structure of transformer to forecast the time series of wind power. Second, we adapt the output of multiple scales by making improvements to the transformer model. Finally, through simulation experiments, the effectiveness of the proposed improved transformer model for multi-scale wind power time series prediction is verified.

Citation Download Citation

Xuefeng Gao, Hao Li, Zhenduo Gao, Xinhong Wang, Yu Shi, and Jia Jia "Prediction of multi-scale wind power time series based on transformer", Proc. SPIE 12598, Eighth International Conference on Energy Materials and Electrical Engineering (ICEMEE 2022), 1259818 (25 April 2023); https://doi.org/10.1117/12.2673035

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available