Deep learning model-transformer based wind power forecasting approach

The uncertainty and fluctuation are the major challenges casted by the large penetration of wind power (WP). As one of the most important solutions for tackling these issues, accurate forecasting is able to enhance the wind energy consumption and improve the penetration rate of WP. In this paper, we...

Full description

Bibliographic Details
Main Authors: Sheng Huang, Chang Yan, Yinpeng Qu
Format: Article
Language:English
Published: Frontiers Media S.A. 2023-01-01
Series:Frontiers in Energy Research
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fenrg.2022.1055683/full
_version_ 1797952139573592064
author Sheng Huang
Chang Yan
Yinpeng Qu
author_facet Sheng Huang
Chang Yan
Yinpeng Qu
author_sort Sheng Huang
collection DOAJ
description The uncertainty and fluctuation are the major challenges casted by the large penetration of wind power (WP). As one of the most important solutions for tackling these issues, accurate forecasting is able to enhance the wind energy consumption and improve the penetration rate of WP. In this paper, we propose a deep learning model-transformer based wind power forecasting (WPF) model. The transformer is a neural network architecture based on the attention mechanism, which is clearly different from other deep learning models such as CNN or RNN. The basic unit of the transformer network consists of residual structure, self-attention mechanism and feedforward network. The overall multilayer encoder to decoder structure enables the network to complete modeling of sequential data. By comparing the forecasting results with other four deep learning models, such as LSTM, the accuracy and efficiency of transformer have been validated. Furthermore, the migration learning experiments show that transformer can also provide good migration performance.
first_indexed 2024-04-10T22:41:35Z
format Article
id doaj.art-530d514d3d91450db4d8343f2b944c15
institution Directory Open Access Journal
issn 2296-598X
language English
last_indexed 2024-04-10T22:41:35Z
publishDate 2023-01-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Energy Research
spelling doaj.art-530d514d3d91450db4d8343f2b944c152023-01-16T04:21:54ZengFrontiers Media S.A.Frontiers in Energy Research2296-598X2023-01-011010.3389/fenrg.2022.10556831055683Deep learning model-transformer based wind power forecasting approachSheng HuangChang YanYinpeng QuThe uncertainty and fluctuation are the major challenges casted by the large penetration of wind power (WP). As one of the most important solutions for tackling these issues, accurate forecasting is able to enhance the wind energy consumption and improve the penetration rate of WP. In this paper, we propose a deep learning model-transformer based wind power forecasting (WPF) model. The transformer is a neural network architecture based on the attention mechanism, which is clearly different from other deep learning models such as CNN or RNN. The basic unit of the transformer network consists of residual structure, self-attention mechanism and feedforward network. The overall multilayer encoder to decoder structure enables the network to complete modeling of sequential data. By comparing the forecasting results with other four deep learning models, such as LSTM, the accuracy and efficiency of transformer have been validated. Furthermore, the migration learning experiments show that transformer can also provide good migration performance.https://www.frontiersin.org/articles/10.3389/fenrg.2022.1055683/fullwind power forecastingtransformerdeep learningdata drivenattention mechanism
spellingShingle Sheng Huang
Chang Yan
Yinpeng Qu
Deep learning model-transformer based wind power forecasting approach
Frontiers in Energy Research
wind power forecasting
transformer
deep learning
data driven
attention mechanism
title Deep learning model-transformer based wind power forecasting approach
title_full Deep learning model-transformer based wind power forecasting approach
title_fullStr Deep learning model-transformer based wind power forecasting approach
title_full_unstemmed Deep learning model-transformer based wind power forecasting approach
title_short Deep learning model-transformer based wind power forecasting approach
title_sort deep learning model transformer based wind power forecasting approach
topic wind power forecasting
transformer
deep learning
data driven
attention mechanism
url https://www.frontiersin.org/articles/10.3389/fenrg.2022.1055683/full
work_keys_str_mv AT shenghuang deeplearningmodeltransformerbasedwindpowerforecastingapproach
AT changyan deeplearningmodeltransformerbasedwindpowerforecastingapproach
AT yinpengqu deeplearningmodeltransformerbasedwindpowerforecastingapproach