Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network

Anto Ryu, Masakazu Ito, Hideo Ishii, Yasuhiro Hayashi

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

Abstract

The installation of photovoltaic system (PV) is increasing rapidly across the world. However, the fluctuation of PV output causes serious challenges in the power grid operation. Among the fluctuation, a quick part of fluctuation is mainly caused by the change of cloud coverage. A total-sky imager (TSI), measuring device to take sky and cloud, could be useful for short-term solar irradiance forecasting. In this paper, a convolutional neural network (CNN) is applied to forecasting model (called CNN model) to forecast 5-20 min ahead of global horizontal irradiance (GHI) using total-sky images and lagged GHI. To verify the effectiveness of CNN, three forecasting models are compared. They are the persistence model, the CNN model using only total-sky images, and the CNN model using both total-sky images and lagged GHI. From the computation, the proposed CNN model using both total-sky images and lagged GHI performs root-mean-square error (RMSE) of 49-177W/m2, 93-146W/m2, 71-118W/m2 in sunny day, partly cloudy day and overcast day, respectively. From these results, the proposed method is shown to be suitable for short-term solar irradiance forecasting.

Original languageEnglish
Title of host publication2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages627-631
Number of pages5
ISBN (Electronic)9781538674345
DOIs
Publication statusPublished - 2019 May 15
Event2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019 - Bangkok, Thailand
Duration: 2019 Mar 192019 Mar 23

Publication series

Name2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019

Conference

Conference2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019
CountryThailand
CityBangkok
Period19/3/1919/3/23

Fingerprint

Image sensors
Neural networks
Mean square error

Keywords

  • Convolutional neural network
  • short-term forecasting
  • Solar irradiance forecasting
  • total-sky imager

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Cite this

Ryu, A., Ito, M., Ishii, H., & Hayashi, Y. (2019). Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network. In 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019 (pp. 627-631). [8715984] (2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GTDAsia.2019.8715984

Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network. / Ryu, Anto; Ito, Masakazu; Ishii, Hideo; Hayashi, Yasuhiro.

2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 627-631 8715984 (2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019).

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

Ryu, A, Ito, M, Ishii, H & Hayashi, Y 2019, Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network. in 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019., 8715984, 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019, Institute of Electrical and Electronics Engineers Inc., pp. 627-631, 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019, Bangkok, Thailand, 19/3/19. https://doi.org/10.1109/GTDAsia.2019.8715984
Ryu A, Ito M, Ishii H, Hayashi Y. Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network. In 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 627-631. 8715984. (2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019). https://doi.org/10.1109/GTDAsia.2019.8715984
Ryu, Anto ; Ito, Masakazu ; Ishii, Hideo ; Hayashi, Yasuhiro. / Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network. 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 627-631 (2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019).
@inproceedings{780d8f9231654aeeabbb742cc3591f2c,
title = "Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network",
abstract = "The installation of photovoltaic system (PV) is increasing rapidly across the world. However, the fluctuation of PV output causes serious challenges in the power grid operation. Among the fluctuation, a quick part of fluctuation is mainly caused by the change of cloud coverage. A total-sky imager (TSI), measuring device to take sky and cloud, could be useful for short-term solar irradiance forecasting. In this paper, a convolutional neural network (CNN) is applied to forecasting model (called CNN model) to forecast 5-20 min ahead of global horizontal irradiance (GHI) using total-sky images and lagged GHI. To verify the effectiveness of CNN, three forecasting models are compared. They are the persistence model, the CNN model using only total-sky images, and the CNN model using both total-sky images and lagged GHI. From the computation, the proposed CNN model using both total-sky images and lagged GHI performs root-mean-square error (RMSE) of 49-177W/m2, 93-146W/m2, 71-118W/m2 in sunny day, partly cloudy day and overcast day, respectively. From these results, the proposed method is shown to be suitable for short-term solar irradiance forecasting.",
keywords = "Convolutional neural network, short-term forecasting, Solar irradiance forecasting, total-sky imager",
author = "Anto Ryu and Masakazu Ito and Hideo Ishii and Yasuhiro Hayashi",
year = "2019",
month = "5",
day = "15",
doi = "10.1109/GTDAsia.2019.8715984",
language = "English",
series = "2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "627--631",
booktitle = "2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019",

}

TY - GEN

T1 - Preliminary Analysis of Short-term Solar Irradiance Forecasting by using Total-sky Imager and Convolutional Neural Network

AU - Ryu, Anto

AU - Ito, Masakazu

AU - Ishii, Hideo

AU - Hayashi, Yasuhiro

PY - 2019/5/15

Y1 - 2019/5/15

N2 - The installation of photovoltaic system (PV) is increasing rapidly across the world. However, the fluctuation of PV output causes serious challenges in the power grid operation. Among the fluctuation, a quick part of fluctuation is mainly caused by the change of cloud coverage. A total-sky imager (TSI), measuring device to take sky and cloud, could be useful for short-term solar irradiance forecasting. In this paper, a convolutional neural network (CNN) is applied to forecasting model (called CNN model) to forecast 5-20 min ahead of global horizontal irradiance (GHI) using total-sky images and lagged GHI. To verify the effectiveness of CNN, three forecasting models are compared. They are the persistence model, the CNN model using only total-sky images, and the CNN model using both total-sky images and lagged GHI. From the computation, the proposed CNN model using both total-sky images and lagged GHI performs root-mean-square error (RMSE) of 49-177W/m2, 93-146W/m2, 71-118W/m2 in sunny day, partly cloudy day and overcast day, respectively. From these results, the proposed method is shown to be suitable for short-term solar irradiance forecasting.

AB - The installation of photovoltaic system (PV) is increasing rapidly across the world. However, the fluctuation of PV output causes serious challenges in the power grid operation. Among the fluctuation, a quick part of fluctuation is mainly caused by the change of cloud coverage. A total-sky imager (TSI), measuring device to take sky and cloud, could be useful for short-term solar irradiance forecasting. In this paper, a convolutional neural network (CNN) is applied to forecasting model (called CNN model) to forecast 5-20 min ahead of global horizontal irradiance (GHI) using total-sky images and lagged GHI. To verify the effectiveness of CNN, three forecasting models are compared. They are the persistence model, the CNN model using only total-sky images, and the CNN model using both total-sky images and lagged GHI. From the computation, the proposed CNN model using both total-sky images and lagged GHI performs root-mean-square error (RMSE) of 49-177W/m2, 93-146W/m2, 71-118W/m2 in sunny day, partly cloudy day and overcast day, respectively. From these results, the proposed method is shown to be suitable for short-term solar irradiance forecasting.

KW - Convolutional neural network

KW - short-term forecasting

KW - Solar irradiance forecasting

KW - total-sky imager

UR - http://www.scopus.com/inward/record.url?scp=85067038761&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067038761&partnerID=8YFLogxK

U2 - 10.1109/GTDAsia.2019.8715984

DO - 10.1109/GTDAsia.2019.8715984

M3 - Conference contribution

AN - SCOPUS:85067038761

T3 - 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019

SP - 627

EP - 631

BT - 2019 IEEE PES GTD Grand International Conference and Exposition Asia, GTD Asia 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -