A stochastic framework for solar irradiance forecasting using condition random field

Jin Xu; Shinjae Yoo; Dantong Yu; Hao Huang; Dong Huang; John Heiser; Paul Kalb

doi:10.1007/978-3-319-18038-0_40

A stochastic framework for solar irradiance forecasting using condition random field

Jin Xu, Shinjae Yoo, Dantong Yu, Hao Huang, Dong Huang, John Heiser, Paul Kalb

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Scopus citations

Abstract

Solar irradiance volatility is a major concern in integrating solar energy micro-grids to the mainstream energy power grid. Accounting for such fluctuations is challenging even with supplier coordination and smart-grid structure implementation. Short-term solar irradiance forecasting is one of the crucial components for maintaining a constant and reliable power output. We propose a novel stochastic solar prediction framework using Conditional Random Fields. The proposed model utilizes features extracted from both cloud images taken by Total Sky Imagers and historical statistics to synergistically reduce the prediction error by 25-40% in terms of MAE in 1-5 minute forecast experiments over the baseline methods.

Original language	English (US)
Title of host publication	Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings
Editors	Tu-Bao Ho, Hiroshi Motoda, Hiroshi Motoda, Ee-Peng Lim, Tru Cao, David Cheung, Zhi-Hua Zhou
Publisher	Springer Verlag
Pages	511-524
Number of pages	14
ISBN (Print)	9783319180373
DOIs	https://doi.org/10.1007/978-3-319-18038-0_40
State	Published - 2015
Externally published	Yes
Event	19th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2015 - Ho Chi Minh City, Viet Nam Duration: May 19 2015 → May 22 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9077
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	19th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2015
Country/Territory	Viet Nam
City	Ho Chi Minh City
Period	5/19/15 → 5/22/15

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Keywords

Conditional random field
Solar forecasting
Stochastic model

Access to Document

10.1007/978-3-319-18038-0_40

Cite this

Xu, J., Yoo, S., Yu, D., Huang, H., Huang, D., Heiser, J., & Kalb, P. (2015). A stochastic framework for solar irradiance forecasting using condition random field. In T.-B. Ho, H. Motoda, H. Motoda, E.-P. Lim, T. Cao, D. Cheung, & Z.-H. Zhou (Eds.), Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings (pp. 511-524). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9077). Springer Verlag. https://doi.org/10.1007/978-3-319-18038-0_40

Xu, Jin ; Yoo, Shinjae ; Yu, Dantong et al. / A stochastic framework for solar irradiance forecasting using condition random field. Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings. editor / Tu-Bao Ho ; Hiroshi Motoda ; Hiroshi Motoda ; Ee-Peng Lim ; Tru Cao ; David Cheung ; Zhi-Hua Zhou. Springer Verlag, 2015. pp. 511-524 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ec448d47906b44a3a162e444229f4aa1,

title = "A stochastic framework for solar irradiance forecasting using condition random field",

abstract = "Solar irradiance volatility is a major concern in integrating solar energy micro-grids to the mainstream energy power grid. Accounting for such fluctuations is challenging even with supplier coordination and smart-grid structure implementation. Short-term solar irradiance forecasting is one of the crucial components for maintaining a constant and reliable power output. We propose a novel stochastic solar prediction framework using Conditional Random Fields. The proposed model utilizes features extracted from both cloud images taken by Total Sky Imagers and historical statistics to synergistically reduce the prediction error by 25-40% in terms of MAE in 1-5 minute forecast experiments over the baseline methods.",

keywords = "Conditional random field, Solar forecasting, Stochastic model",

author = "Jin Xu and Shinjae Yoo and Dantong Yu and Hao Huang and Dong Huang and John Heiser and Paul Kalb",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 19th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2015 ; Conference date: 19-05-2015 Through 22-05-2015",

year = "2015",

doi = "10.1007/978-3-319-18038-0_40",

language = "English (US)",

isbn = "9783319180373",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "511--524",

editor = "Tu-Bao Ho and Hiroshi Motoda and Hiroshi Motoda and Ee-Peng Lim and Tru Cao and David Cheung and Zhi-Hua Zhou",

booktitle = "Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings",

address = "Germany",

}

Xu, J, Yoo, S, Yu, D, Huang, H, Huang, D, Heiser, J & Kalb, P 2015, A stochastic framework for solar irradiance forecasting using condition random field. in T-B Ho, H Motoda, H Motoda, E-P Lim, T Cao, D Cheung & Z-H Zhou (eds), Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9077, Springer Verlag, pp. 511-524, 19th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2015, Ho Chi Minh City, Viet Nam, 5/19/15. https://doi.org/10.1007/978-3-319-18038-0_40

A stochastic framework for solar irradiance forecasting using condition random field. / Xu, Jin; Yoo, Shinjae; Yu, Dantong et al.
Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings. ed. / Tu-Bao Ho; Hiroshi Motoda; Hiroshi Motoda; Ee-Peng Lim; Tru Cao; David Cheung; Zhi-Hua Zhou. Springer Verlag, 2015. p. 511-524 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9077).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A stochastic framework for solar irradiance forecasting using condition random field

AU - Xu, Jin

AU - Yoo, Shinjae

AU - Yu, Dantong

AU - Huang, Hao

AU - Huang, Dong

AU - Heiser, John

AU - Kalb, Paul

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - Solar irradiance volatility is a major concern in integrating solar energy micro-grids to the mainstream energy power grid. Accounting for such fluctuations is challenging even with supplier coordination and smart-grid structure implementation. Short-term solar irradiance forecasting is one of the crucial components for maintaining a constant and reliable power output. We propose a novel stochastic solar prediction framework using Conditional Random Fields. The proposed model utilizes features extracted from both cloud images taken by Total Sky Imagers and historical statistics to synergistically reduce the prediction error by 25-40% in terms of MAE in 1-5 minute forecast experiments over the baseline methods.

AB - Solar irradiance volatility is a major concern in integrating solar energy micro-grids to the mainstream energy power grid. Accounting for such fluctuations is challenging even with supplier coordination and smart-grid structure implementation. Short-term solar irradiance forecasting is one of the crucial components for maintaining a constant and reliable power output. We propose a novel stochastic solar prediction framework using Conditional Random Fields. The proposed model utilizes features extracted from both cloud images taken by Total Sky Imagers and historical statistics to synergistically reduce the prediction error by 25-40% in terms of MAE in 1-5 minute forecast experiments over the baseline methods.

KW - Conditional random field

KW - Solar forecasting

KW - Stochastic model

UR - http://www.scopus.com/inward/record.url?scp=84945931513&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84945931513&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-18038-0_40

DO - 10.1007/978-3-319-18038-0_40

M3 - Conference contribution

AN - SCOPUS:84945931513

SN - 9783319180373

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 511

EP - 524

BT - Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings

A2 - Ho, Tu-Bao

A2 - Motoda, Hiroshi

A2 - Lim, Ee-Peng

A2 - Cao, Tru

A2 - Cheung, David

A2 - Zhou, Zhi-Hua

PB - Springer Verlag

T2 - 19th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2015

Y2 - 19 May 2015 through 22 May 2015

ER -

Xu J, Yoo S, Yu D, Huang H, Huang D, Heiser J et al. A stochastic framework for solar irradiance forecasting using condition random field. In Ho TB, Motoda H, Motoda H, Lim EP, Cao T, Cheung D, Zhou ZH, editors, Advances in Knowledge Discovery and Data Mining - 19th Pacific-Asia Conference, PAKDD 2015, Proceedings. Springer Verlag. 2015. p. 511-524. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-18038-0_40

A stochastic framework for solar irradiance forecasting using condition random field

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this