LASSO vector autoregression structures for very short-term wind power forecasting

L. Cavalcante, Ricardo J. Bessa, Marisa Reis, Jethro Browell

Research output: Contribution to journalArticle

Abstract

The deployment of smart grids and renewable energy dispatch centers motivates the development of forecasting techniques that take advantage of near real-time measurements collected from geographically distributed sensors. This paper describes a forecasting methodology that explores a set of different sparse structures for the vector autoregression (VAR) model using the Least Absolute Shrinkage and Selection Operator (LASSO) framework. The alternating direction method of multipliers is applied to fit the different VAR-LASSO variants and create a scalable forecasting method supported by parallel computing and fast convergence, which can be used by system operators and renewable power plant operators. A test case with 66 wind power plants is used to show the improvement in forecasting skill from exploring distributed sparse structures. The proposed solution outperformed the conventional autoregressive and vector autoregressive models, as well as a sparse-VAR model from the state of the art.LASSO Vector Autoregression Structures for Very Short-term Wind Power Forecasting
LanguageEnglish
Number of pages24
JournalWind Energy
Publication statusAccepted/In press - 23 Aug 2016

Fingerprint

Wind power
Power plants
Parallel processing systems
Time measurement
Sensors

Keywords

  • wind power
  • vector autoregression
  • scalability
  • sparse
  • renewable energy
  • parallel computing

Cite this

@article{13c7e71d05c34fe5ace95f5139fb1548,
title = "LASSO vector autoregression structures for very short-term wind power forecasting",
abstract = "The deployment of smart grids and renewable energy dispatch centers motivates the development of forecasting techniques that take advantage of near real-time measurements collected from geographically distributed sensors. This paper describes a forecasting methodology that explores a set of different sparse structures for the vector autoregression (VAR) model using the Least Absolute Shrinkage and Selection Operator (LASSO) framework. The alternating direction method of multipliers is applied to fit the different VAR-LASSO variants and create a scalable forecasting method supported by parallel computing and fast convergence, which can be used by system operators and renewable power plant operators. A test case with 66 wind power plants is used to show the improvement in forecasting skill from exploring distributed sparse structures. The proposed solution outperformed the conventional autoregressive and vector autoregressive models, as well as a sparse-VAR model from the state of the art.LASSO Vector Autoregression Structures for Very Short-term Wind Power Forecasting",
keywords = "wind power, vector autoregression, scalability, sparse, renewable energy, parallel computing",
author = "L. Cavalcante and Bessa, {Ricardo J.} and Marisa Reis and Jethro Browell",
note = "This is the peer reviewed version of the following article: Cavalcante, L., Bessa, R. J., Reis, M., & Browell, J. (2016). LASSO vector autoregression structures for very short-term wind power forecasting. Wind Energy, which has been published in final form at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1099-1824. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving.",
year = "2016",
month = "8",
day = "23",
language = "English",
journal = "Wind Energy",
issn = "1095-4244",

}

LASSO vector autoregression structures for very short-term wind power forecasting. / Cavalcante, L.; Bessa, Ricardo J.; Reis, Marisa; Browell, Jethro.

In: Wind Energy, 23.08.2016.

Research output: Contribution to journalArticle

TY - JOUR

T1 - LASSO vector autoregression structures for very short-term wind power forecasting

AU - Cavalcante, L.

AU - Bessa, Ricardo J.

AU - Reis, Marisa

AU - Browell, Jethro

N1 - This is the peer reviewed version of the following article: Cavalcante, L., Bessa, R. J., Reis, M., & Browell, J. (2016). LASSO vector autoregression structures for very short-term wind power forecasting. Wind Energy, which has been published in final form at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1099-1824. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving.

PY - 2016/8/23

Y1 - 2016/8/23

N2 - The deployment of smart grids and renewable energy dispatch centers motivates the development of forecasting techniques that take advantage of near real-time measurements collected from geographically distributed sensors. This paper describes a forecasting methodology that explores a set of different sparse structures for the vector autoregression (VAR) model using the Least Absolute Shrinkage and Selection Operator (LASSO) framework. The alternating direction method of multipliers is applied to fit the different VAR-LASSO variants and create a scalable forecasting method supported by parallel computing and fast convergence, which can be used by system operators and renewable power plant operators. A test case with 66 wind power plants is used to show the improvement in forecasting skill from exploring distributed sparse structures. The proposed solution outperformed the conventional autoregressive and vector autoregressive models, as well as a sparse-VAR model from the state of the art.LASSO Vector Autoregression Structures for Very Short-term Wind Power Forecasting

AB - The deployment of smart grids and renewable energy dispatch centers motivates the development of forecasting techniques that take advantage of near real-time measurements collected from geographically distributed sensors. This paper describes a forecasting methodology that explores a set of different sparse structures for the vector autoregression (VAR) model using the Least Absolute Shrinkage and Selection Operator (LASSO) framework. The alternating direction method of multipliers is applied to fit the different VAR-LASSO variants and create a scalable forecasting method supported by parallel computing and fast convergence, which can be used by system operators and renewable power plant operators. A test case with 66 wind power plants is used to show the improvement in forecasting skill from exploring distributed sparse structures. The proposed solution outperformed the conventional autoregressive and vector autoregressive models, as well as a sparse-VAR model from the state of the art.LASSO Vector Autoregression Structures for Very Short-term Wind Power Forecasting

KW - wind power

KW - vector autoregression

KW - scalability

KW - sparse

KW - renewable energy

KW - parallel computing

UR - http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1099-1824

M3 - Article

JO - Wind Energy

T2 - Wind Energy

JF - Wind Energy

SN - 1095-4244

ER -