A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning

Li, Qiang; Liu, Yongzhi; Yan, Xinyue; Zhang, Haipeng; Wang, Siyu; Li, Ran

doi:10.3390/en19020349

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning

by

Qiang Li

¹,

Yongzhi Liu

¹,

Xinyue Yan

²,

Haipeng Zhang

^3,4,

Siyu Wang

¹ and

Ran Li

^2,3,4,*

¹

Inner Mongolia Power (Group) Co., Ltd., Hohhot 010010, China

²

College of Smart Energy, Shanghai Jiao Tong University, Shanghai 200240, China

³

Key Laboratory of Control of Power Transmission and Conversion, Ministry of Education, Shanghai Jiao Tong University, Shanghai 200240, China

⁴

Shanghai Non-Carbon Energy Conversion and Utilization Institute, Shanghai 200240, China

^*

Author to whom correspondence should be addressed.

Energies 2026, 19(2), 349; https://doi.org/10.3390/en19020349 (registering DOI)

Submission received: 1 December 2025 / Revised: 27 December 2025 / Accepted: 29 December 2025 / Published: 10 January 2026

(This article belongs to the Special Issue Challenges and Research Trends of Integrated Zero-Carbon Power Plant)

Download Versions Notes

Abstract

In short-term power forecasting for wind farms, factors such as weather conditions and geographic location lead to certain correlations in the power output of different wind farms, resulting in complex coupling relationships between them. Traditional wind power forecasting methods often predict each wind farm independently, without considering these coupling relationships. To address this issue, this paper proposes a multi-task Transformer model based on multiple decoders, which accounts for the intrinsic connections between different wind farms, enabling joint power forecasting across multiple sites. The proposed model adopts a single encoder-multiple decoder structure, where a unified encoder processes all input data, and multiple decoders perform prediction tasks for each wind farm separately. Testing on actual wind farm data from the Inner Mongolia region of China shows that, compared to other forecasting models, the proposed model significantly improves the accuracy of power predictions for different wind farms.

Keywords: renewable energy forecasting; multi-decoder; Transformer model; attention mechanism; deep learning

Share and Cite

MDPI and ACS Style

Li, Q.; Liu, Y.; Yan, X.; Zhang, H.; Wang, S.; Li, R. A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning. Energies 2026, 19, 349. https://doi.org/10.3390/en19020349

AMA Style

Li Q, Liu Y, Yan X, Zhang H, Wang S, Li R. A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning. Energies. 2026; 19(2):349. https://doi.org/10.3390/en19020349

Chicago/Turabian Style

Li, Qiang, Yongzhi Liu, Xinyue Yan, Haipeng Zhang, Siyu Wang, and Ran Li. 2026. "A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning" Energies 19, no. 2: 349. https://doi.org/10.3390/en19020349

APA Style

Li, Q., Liu, Y., Yan, X., Zhang, H., Wang, S., & Li, R. (2026). A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning. Energies, 19(2), 349. https://doi.org/10.3390/en19020349

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

A Short-Term Wind Power Forecasting Method Based on Multi-Decoder and Multi-Task Learning

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI