The all-seeing eye: Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Direct Normal Irradiance (DNI) is a critical component of solar irradiation for estimating Plane of Array (POA) irradiance on flat plate systems as well as for concentrating systems. Current approaches to measuring or estimating DNI suffer from drawbacks of either high equipment costs or low precision. A new approach is proposed, using Artificial Neural Networks to estimate DNI from the irradiance measures of five pyranometers. We consider various neural network topologies and training data sources and analyze the resulting errors. The estimated systems costs are found to be an order of magnitude less than a traditional pyrheliometer setup, with precision and accuracy significantly improved relative to a single pyranometer system.

Original languageEnglish (US)
Title of host publication43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy
PublisherAmerican Solar Energy Society
Pages511-518
Number of pages8
ISBN (Electronic)9781510801790
StatePublished - Jan 1 2014
Event43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy - San Francisco, United States
Duration: Jul 6 2014Jul 10 2014

Publication series

Name43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy
Volume1

Other

Other43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy
CountryUnited States
CitySan Francisco
Period7/6/147/10/14

Fingerprint

Neural networks
Costs
Topology
Irradiation

All Science Journal Classification (ASJC) codes

  • Renewable Energy, Sustainability and the Environment

Cite this

Srikrishnan, V., Brownson, J., & Young, G. S. (2014). The all-seeing eye: Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance. In 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy (pp. 511-518). (43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy; Vol. 1). American Solar Energy Society.
Srikrishnan, Vivek ; Brownson, Jeffrey ; Young, George Spencer. / The all-seeing eye : Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance. 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy. American Solar Energy Society, 2014. pp. 511-518 (43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy).
@inproceedings{2c50ca25ff87484e9fbcc856d2b554ef,
title = "The all-seeing eye: Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance",
abstract = "Direct Normal Irradiance (DNI) is a critical component of solar irradiation for estimating Plane of Array (POA) irradiance on flat plate systems as well as for concentrating systems. Current approaches to measuring or estimating DNI suffer from drawbacks of either high equipment costs or low precision. A new approach is proposed, using Artificial Neural Networks to estimate DNI from the irradiance measures of five pyranometers. We consider various neural network topologies and training data sources and analyze the resulting errors. The estimated systems costs are found to be an order of magnitude less than a traditional pyrheliometer setup, with precision and accuracy significantly improved relative to a single pyranometer system.",
author = "Vivek Srikrishnan and Jeffrey Brownson and Young, {George Spencer}",
year = "2014",
month = "1",
day = "1",
language = "English (US)",
series = "43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy",
publisher = "American Solar Energy Society",
pages = "511--518",
booktitle = "43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy",

}

Srikrishnan, V, Brownson, J & Young, GS 2014, The all-seeing eye: Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance. in 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy. 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy, vol. 1, American Solar Energy Society, pp. 511-518, 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy, San Francisco, United States, 7/6/14.

The all-seeing eye : Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance. / Srikrishnan, Vivek; Brownson, Jeffrey; Young, George Spencer.

43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy. American Solar Energy Society, 2014. p. 511-518 (43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy; Vol. 1).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - The all-seeing eye

T2 - Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance

AU - Srikrishnan, Vivek

AU - Brownson, Jeffrey

AU - Young, George Spencer

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Direct Normal Irradiance (DNI) is a critical component of solar irradiation for estimating Plane of Array (POA) irradiance on flat plate systems as well as for concentrating systems. Current approaches to measuring or estimating DNI suffer from drawbacks of either high equipment costs or low precision. A new approach is proposed, using Artificial Neural Networks to estimate DNI from the irradiance measures of five pyranometers. We consider various neural network topologies and training data sources and analyze the resulting errors. The estimated systems costs are found to be an order of magnitude less than a traditional pyrheliometer setup, with precision and accuracy significantly improved relative to a single pyranometer system.

AB - Direct Normal Irradiance (DNI) is a critical component of solar irradiation for estimating Plane of Array (POA) irradiance on flat plate systems as well as for concentrating systems. Current approaches to measuring or estimating DNI suffer from drawbacks of either high equipment costs or low precision. A new approach is proposed, using Artificial Neural Networks to estimate DNI from the irradiance measures of five pyranometers. We consider various neural network topologies and training data sources and analyze the resulting errors. The estimated systems costs are found to be an order of magnitude less than a traditional pyrheliometer setup, with precision and accuracy significantly improved relative to a single pyranometer system.

UR - http://www.scopus.com/inward/record.url?scp=84944722807&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944722807&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84944722807

T3 - 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy

SP - 511

EP - 518

BT - 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy

PB - American Solar Energy Society

ER -

Srikrishnan V, Brownson J, Young GS. The all-seeing eye: Using multi-pyranometer arrays and neural networks to estimate direct normal irradiance. In 43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy. American Solar Energy Society. 2014. p. 511-518. (43rd ASES National Solar Conference 2014, SOLAR 2014, Including the 39th National Passive Solar Conference and the 2nd Meeting of Young and Emerging Professionals in Renewable Energy).