Wind Park Power Prediction: Attention-Based Graph Networks and Deep Learning to Capture Wake Losses

被引:12
作者
Bentsen, Lars Odegaard [1 ]
Warakagoda, Narada Dilp [1 ]
Stenbro, Roy [2 ]
Engelstad, Paal [1 ]
机构
[1] Univ Oslo, Dept Technol Syst, POB 70, N-2027 Kjeller, Norway
[2] Inst Energy Syst IFE, POB 40, N-2027 Kjeller, Norway
来源
SCIENCE OF MAKING TORQUE FROM WIND, TORQUE 2022 | 2022年 / 2265卷
关键词
NEURAL-NETWORK;
D O I
10.1088/1742-6596/2265/2/022035
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
With the increased penetration of wind energy into the power grid, it has become increasingly important to be able to predict the expected power production for larger wind farms. Deep learning (DL) models can learn complex patterns in the data and have found wide success in predicting wake losses and expected power production. This paper proposes a modular framework for attention-based graph neural networks (GNN), where attention can be applied to any desired component of a graph block. The results show that the model significantly outperforms a multilayer perceptron (MLP) and a bidirectional LSTM (BLSTM) model, while delivering performance on-par with a vanilla GNN model. Moreover, we argue that the proposed graph attention architecture can easily adapt to different applications by offering flexibility into the desired attention operations to be used, which might depend on the specific application. Through analysis of the attention weights, it was showed that employing attention-based GNNs can provide insights into what the models learn. In particular, the attention networks seemed to realise turbine dependencies that aligned with some physical intuition about wake losses.
引用
收藏
页数:11
相关论文
共 33 条
[1]  
Bahdanau D, 2016, Arxiv, DOI arXiv:1409.0473
[2]   A Graph Neural Network Surrogate Model for the Prediction of Turbine Interaction Loss [J].
Bleeg, James .
SCIENCE OF MAKING TORQUE FROM WIND (TORQUE 2020), PTS 1-5, 2020, 1618
[3]  
Fu X, 2019, ICDMW, P149
[4]  
Georgousis S., 2021, IEEE Access
[5]  
Ghaderi A, 2017, Arxiv, DOI arXiv:1707.08110
[6]   A Deep Hybrid Model for Weather Forecasting [J].
Grover, Aditya ;
Kapoor, Ashish ;
Horvitz, Eric .
KDD'15: PROCEEDINGS OF THE 21ST ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2015, :379-386
[7]  
GWEC, 2021, Gwec-global wind report 2021
[8]   A Critical Review of Wind Power Forecasting Methods-Past, Present and Future [J].
Hanifi, Shahram ;
Liu, Xiaolei ;
Lin, Zi ;
Lotfian, Saeid .
ENERGIES, 2020, 13 (15)
[9]  
Hochreiter S, 1997, NEURAL COMPUT, V9, P1735, DOI [10.1162/neco.1997.9.8.1735, 10.1162/neco.1997.9.1.1, 10.1007/978-3-642-24797-2]
[10]   A hybrid deep learning-based neural network for 24-h ahead wind power forecasting [J].
Hong, Ying-Yi ;
Rioflorido, Christian Lian Paulo P. .
APPLIED ENERGY, 2019, 250 :530-539