Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model

Jinhua Zhang, Jie Yan, David Infield, Yongqian Liu, Fue-sang Lien

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Wind power plays a leading role in the development of renewable energy. However, the random nature of wind turbine power and its associated uncertainty create challenges in dispatching this power effectively in the power system, which can result in unnecessary curtailment of the wind turbine power. Improving the accuracy of wind turbine power forecasting is an effective measure for resolving such problems. This study uses a deep learning network to forecast the wind turbine power based on a long short-term memory network (LSTM) algorithm and uses the Gaussian mixture model (GMM) to analyze the error distribution characteristics of short-term wind turbine power forecasting. The LSTM algorithm is used to forecast the power and uncertainties for three wind turbines within a wind farm. According to numerical weather prediction (NWP) data and historical power data for three turbines, the forecasting accuracy of the turbine with the largest number of training samples is the best of the three. For one of the turbines, the LSTM, radial basis function (RBF), wavelet, deep belief network (DBN), back propagation neural networks (BPNN), and Elman neural network (ELMAN) have been used to forecast the wind turbine power. This study compares the results and demonstrates that LSTM can greatly improve the forecasting accuracy. Moreover, this study obtains different confidence intervals for the three units according to the GMM, mixture density neural network (MDN), and relevance vector machine (RVM) model results. The LSTM method is shown to have higher accuracy and faster convergence than the other methods. However, the GMM method has better performance and evaluation than other methods and thus has practical application value for wind turbine power dispatching.
LanguageEnglish
Pages229-244
Number of pages16
JournalApplied Energy
Volume241
Early online date12 Mar 2019
DOIs
Publication statusPublished - 1 May 2019

Fingerprint

Uncertainty analysis
wind turbine
uncertainty analysis
Wind turbines
turbine
Turbines
Neural networks
wind farm
Radial basis function networks
Long short-term memory
back propagation
wind power
Bayesian networks
Backpropagation
Farms
Wind power
confidence interval
wavelet
learning
weather

Keywords

  • long short-term memory network
  • Gaussian mixture model
  • wind turbine power
  • short-term prediction
  • uncertainty analysis

Cite this

@article{f6961740f39d407baf4d716df454197b,
title = "Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model",
abstract = "Wind power plays a leading role in the development of renewable energy. However, the random nature of wind turbine power and its associated uncertainty create challenges in dispatching this power effectively in the power system, which can result in unnecessary curtailment of the wind turbine power. Improving the accuracy of wind turbine power forecasting is an effective measure for resolving such problems. This study uses a deep learning network to forecast the wind turbine power based on a long short-term memory network (LSTM) algorithm and uses the Gaussian mixture model (GMM) to analyze the error distribution characteristics of short-term wind turbine power forecasting. The LSTM algorithm is used to forecast the power and uncertainties for three wind turbines within a wind farm. According to numerical weather prediction (NWP) data and historical power data for three turbines, the forecasting accuracy of the turbine with the largest number of training samples is the best of the three. For one of the turbines, the LSTM, radial basis function (RBF), wavelet, deep belief network (DBN), back propagation neural networks (BPNN), and Elman neural network (ELMAN) have been used to forecast the wind turbine power. This study compares the results and demonstrates that LSTM can greatly improve the forecasting accuracy. Moreover, this study obtains different confidence intervals for the three units according to the GMM, mixture density neural network (MDN), and relevance vector machine (RVM) model results. The LSTM method is shown to have higher accuracy and faster convergence than the other methods. However, the GMM method has better performance and evaluation than other methods and thus has practical application value for wind turbine power dispatching.",
keywords = "long short-term memory network, Gaussian mixture model, wind turbine power, short-term prediction, uncertainty analysis",
author = "Jinhua Zhang and Jie Yan and David Infield and Yongqian Liu and Fue-sang Lien",
year = "2019",
month = "5",
day = "1",
doi = "10.1016/j.apenergy.2019.03.044",
language = "English",
volume = "241",
pages = "229--244",
journal = "Applied Energy",
issn = "0306-2619",

}

Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model. / Zhang, Jinhua; Yan, Jie; Infield, David; Liu, Yongqian; Lien, Fue-sang.

In: Applied Energy, Vol. 241, 01.05.2019, p. 229-244.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model

AU - Zhang, Jinhua

AU - Yan, Jie

AU - Infield, David

AU - Liu, Yongqian

AU - Lien, Fue-sang

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Wind power plays a leading role in the development of renewable energy. However, the random nature of wind turbine power and its associated uncertainty create challenges in dispatching this power effectively in the power system, which can result in unnecessary curtailment of the wind turbine power. Improving the accuracy of wind turbine power forecasting is an effective measure for resolving such problems. This study uses a deep learning network to forecast the wind turbine power based on a long short-term memory network (LSTM) algorithm and uses the Gaussian mixture model (GMM) to analyze the error distribution characteristics of short-term wind turbine power forecasting. The LSTM algorithm is used to forecast the power and uncertainties for three wind turbines within a wind farm. According to numerical weather prediction (NWP) data and historical power data for three turbines, the forecasting accuracy of the turbine with the largest number of training samples is the best of the three. For one of the turbines, the LSTM, radial basis function (RBF), wavelet, deep belief network (DBN), back propagation neural networks (BPNN), and Elman neural network (ELMAN) have been used to forecast the wind turbine power. This study compares the results and demonstrates that LSTM can greatly improve the forecasting accuracy. Moreover, this study obtains different confidence intervals for the three units according to the GMM, mixture density neural network (MDN), and relevance vector machine (RVM) model results. The LSTM method is shown to have higher accuracy and faster convergence than the other methods. However, the GMM method has better performance and evaluation than other methods and thus has practical application value for wind turbine power dispatching.

AB - Wind power plays a leading role in the development of renewable energy. However, the random nature of wind turbine power and its associated uncertainty create challenges in dispatching this power effectively in the power system, which can result in unnecessary curtailment of the wind turbine power. Improving the accuracy of wind turbine power forecasting is an effective measure for resolving such problems. This study uses a deep learning network to forecast the wind turbine power based on a long short-term memory network (LSTM) algorithm and uses the Gaussian mixture model (GMM) to analyze the error distribution characteristics of short-term wind turbine power forecasting. The LSTM algorithm is used to forecast the power and uncertainties for three wind turbines within a wind farm. According to numerical weather prediction (NWP) data and historical power data for three turbines, the forecasting accuracy of the turbine with the largest number of training samples is the best of the three. For one of the turbines, the LSTM, radial basis function (RBF), wavelet, deep belief network (DBN), back propagation neural networks (BPNN), and Elman neural network (ELMAN) have been used to forecast the wind turbine power. This study compares the results and demonstrates that LSTM can greatly improve the forecasting accuracy. Moreover, this study obtains different confidence intervals for the three units according to the GMM, mixture density neural network (MDN), and relevance vector machine (RVM) model results. The LSTM method is shown to have higher accuracy and faster convergence than the other methods. However, the GMM method has better performance and evaluation than other methods and thus has practical application value for wind turbine power dispatching.

KW - long short-term memory network

KW - Gaussian mixture model

KW - wind turbine power

KW - short-term prediction

KW - uncertainty analysis

UR - https://www.journals.elsevier.com/applied-energy

U2 - 10.1016/j.apenergy.2019.03.044

DO - 10.1016/j.apenergy.2019.03.044

M3 - Article

VL - 241

SP - 229

EP - 244

JO - Applied Energy

T2 - Applied Energy

JF - Applied Energy

SN - 0306-2619

ER -