Abstract

Wind ramps introduce significant uncertainty into wind power generation. Reliable system operation, however, requires accurate detection and forecast of wind ramps, especially at high penetration levels. In this paper, to deal with the wind ramp dynamics, a support vector machine (SVM)-enhanced Markov model is developed for short-term wind power forecast, based on one key observation from the measurement data that wind ramps often occur with specific patterns. Specifically, using the historical data of the wind turbine power outputs recorded at an actual wind farm, data analytics-based finite-state Markov models are first developed to model the normal fluctuations of wind generation, while taking into account the diurnal nonstationarity and the seasonality of wind generation. Next, the forecast by the SVM is integrated cohesively into the finite-state Markov models. Based on the SVM-enhanced Markov model, both short-term distributional forecasts and point forecasts are then derived. Numerical test results, using real wind generation data traces, demonstrate the significantly improved accuracy of the proposed forecast approach.

Original languageEnglish (US)
Article number7081774
Pages (from-to)791-799
Number of pages9
JournalIEEE Transactions on Sustainable Energy
Volume6
Issue number3
DOIs
StatePublished - Jul 1 2015

Fingerprint

Wind power
Support vector machines
Wind turbines
Farms
Power generation

Keywords

  • Distributional forecast
  • Markov chain
  • point forecast
  • short-term wind power forecast
  • support vector machine (SVM)
  • wind farm

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment

Cite this

Support-vector-machine-enhanced markov model for short-term wind power forecast. / Yang, Lei; He, Miao; Zhang, Junshan; Vittal, Vijay.

In: IEEE Transactions on Sustainable Energy, Vol. 6, No. 3, 7081774, 01.07.2015, p. 791-799.

Research output: Contribution to journalArticle

@article{6db3ee4fd862400e9601ffa9d36c3336,
title = "Support-vector-machine-enhanced markov model for short-term wind power forecast",
abstract = "Wind ramps introduce significant uncertainty into wind power generation. Reliable system operation, however, requires accurate detection and forecast of wind ramps, especially at high penetration levels. In this paper, to deal with the wind ramp dynamics, a support vector machine (SVM)-enhanced Markov model is developed for short-term wind power forecast, based on one key observation from the measurement data that wind ramps often occur with specific patterns. Specifically, using the historical data of the wind turbine power outputs recorded at an actual wind farm, data analytics-based finite-state Markov models are first developed to model the normal fluctuations of wind generation, while taking into account the diurnal nonstationarity and the seasonality of wind generation. Next, the forecast by the SVM is integrated cohesively into the finite-state Markov models. Based on the SVM-enhanced Markov model, both short-term distributional forecasts and point forecasts are then derived. Numerical test results, using real wind generation data traces, demonstrate the significantly improved accuracy of the proposed forecast approach.",
keywords = "Distributional forecast, Markov chain, point forecast, short-term wind power forecast, support vector machine (SVM), wind farm",
author = "Lei Yang and Miao He and Junshan Zhang and Vijay Vittal",
year = "2015",
month = "7",
day = "1",
doi = "10.1109/TSTE.2015.2406814",
language = "English (US)",
volume = "6",
pages = "791--799",
journal = "IEEE Transactions on Sustainable Energy",
issn = "1949-3029",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Support-vector-machine-enhanced markov model for short-term wind power forecast

AU - Yang, Lei

AU - He, Miao

AU - Zhang, Junshan

AU - Vittal, Vijay

PY - 2015/7/1

Y1 - 2015/7/1

N2 - Wind ramps introduce significant uncertainty into wind power generation. Reliable system operation, however, requires accurate detection and forecast of wind ramps, especially at high penetration levels. In this paper, to deal with the wind ramp dynamics, a support vector machine (SVM)-enhanced Markov model is developed for short-term wind power forecast, based on one key observation from the measurement data that wind ramps often occur with specific patterns. Specifically, using the historical data of the wind turbine power outputs recorded at an actual wind farm, data analytics-based finite-state Markov models are first developed to model the normal fluctuations of wind generation, while taking into account the diurnal nonstationarity and the seasonality of wind generation. Next, the forecast by the SVM is integrated cohesively into the finite-state Markov models. Based on the SVM-enhanced Markov model, both short-term distributional forecasts and point forecasts are then derived. Numerical test results, using real wind generation data traces, demonstrate the significantly improved accuracy of the proposed forecast approach.

AB - Wind ramps introduce significant uncertainty into wind power generation. Reliable system operation, however, requires accurate detection and forecast of wind ramps, especially at high penetration levels. In this paper, to deal with the wind ramp dynamics, a support vector machine (SVM)-enhanced Markov model is developed for short-term wind power forecast, based on one key observation from the measurement data that wind ramps often occur with specific patterns. Specifically, using the historical data of the wind turbine power outputs recorded at an actual wind farm, data analytics-based finite-state Markov models are first developed to model the normal fluctuations of wind generation, while taking into account the diurnal nonstationarity and the seasonality of wind generation. Next, the forecast by the SVM is integrated cohesively into the finite-state Markov models. Based on the SVM-enhanced Markov model, both short-term distributional forecasts and point forecasts are then derived. Numerical test results, using real wind generation data traces, demonstrate the significantly improved accuracy of the proposed forecast approach.

KW - Distributional forecast

KW - Markov chain

KW - point forecast

KW - short-term wind power forecast

KW - support vector machine (SVM)

KW - wind farm

UR - http://www.scopus.com/inward/record.url?scp=85027956105&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027956105&partnerID=8YFLogxK

U2 - 10.1109/TSTE.2015.2406814

DO - 10.1109/TSTE.2015.2406814

M3 - Article

VL - 6

SP - 791

EP - 799

JO - IEEE Transactions on Sustainable Energy

JF - IEEE Transactions on Sustainable Energy

SN - 1949-3029

IS - 3

M1 - 7081774

ER -