Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks

Tsai Ping Chu; Jian Hua Jhou; Yih Guang Leu

doi:10.1109/ICSSE50014.2020.9219301

Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks

Tsai Ping Chu, Jian Hua Jhou, Yih Guang Leu^*

^*Corresponding author for this work

Department of Electrical Engineering

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

9 Citations (Scopus)

Abstract

The solar power variability is due to the variability of solar irradiance. Several factors are involved in the situation, such as cloud thickness and air pollution. In this paper, we attempt to find a novel way to predict the amount of solar irradiance. A image-based forecasting method is developed, and Long Short-Term Memory (LSTM) neural network is applied for data training. Daily solar irradiance and sky images are record by the record system, and uploaded to the MySQL database for storage. Feature values obtained by analyzing sky images are used as the input of neural network with solar irradiance. After some performance evaluation indicators were demonstrated, we found that the proposed method has good predictive performance with 5 to 60 minutes in present.

Original language	English
Title of host publication	2020 International Conference on System Science and Engineering, ICSSE 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728159607
DOIs	https://doi.org/10.1109/ICSSE50014.2020.9219301
Publication status	Published - 2020 Aug
Event	2020 International Conference on System Science and Engineering, ICSSE 2020 - Kagawa, Japan Duration: 2020 Aug 31 → 2020 Sept 3

Publication series

Name	2020 International Conference on System Science and Engineering, ICSSE 2020

Conference

Conference	2020 International Conference on System Science and Engineering, ICSSE 2020
Country/Territory	Japan
City	Kagawa
Period	2020/08/31 → 2020/09/03

Keywords

Deep Learning
Image Feature
Recurrent Neural Network
Solar Irradiance Forecasting

ASJC Scopus subject areas

Artificial Intelligence
Energy Engineering and Power Technology
Renewable Energy, Sustainability and the Environment
Civil and Structural Engineering
Electrical and Electronic Engineering
Control and Optimization

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ICSSE50014.2020.9219301

Cite this

Chu, T. P., Jhou, J. H., & Leu, Y. G. (2020). Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks. In 2020 International Conference on System Science and Engineering, ICSSE 2020 Article 9219301 (2020 International Conference on System Science and Engineering, ICSSE 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICSSE50014.2020.9219301

Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks. / Chu, Tsai Ping; Jhou, Jian Hua; Leu, Yih Guang.
2020 International Conference on System Science and Engineering, ICSSE 2020. Institute of Electrical and Electronics Engineers Inc., 2020. 9219301 (2020 International Conference on System Science and Engineering, ICSSE 2020).

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

Chu, TP, Jhou, JH & Leu, YG 2020, Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks. in 2020 International Conference on System Science and Engineering, ICSSE 2020., 9219301, 2020 International Conference on System Science and Engineering, ICSSE 2020, Institute of Electrical and Electronics Engineers Inc., 2020 International Conference on System Science and Engineering, ICSSE 2020, Kagawa, Japan, 2020/08/31. https://doi.org/10.1109/ICSSE50014.2020.9219301

@inproceedings{ab10e9ba41f3446caec559d80f941161,

title = "Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks",

abstract = "The solar power variability is due to the variability of solar irradiance. Several factors are involved in the situation, such as cloud thickness and air pollution. In this paper, we attempt to find a novel way to predict the amount of solar irradiance. A image-based forecasting method is developed, and Long Short-Term Memory (LSTM) neural network is applied for data training. Daily solar irradiance and sky images are record by the record system, and uploaded to the MySQL database for storage. Feature values obtained by analyzing sky images are used as the input of neural network with solar irradiance. After some performance evaluation indicators were demonstrated, we found that the proposed method has good predictive performance with 5 to 60 minutes in present.",

keywords = "Deep Learning, Image Feature, Recurrent Neural Network, Solar Irradiance Forecasting",

author = "Chu, {Tsai Ping} and Jhou, {Jian Hua} and Leu, {Yih Guang}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 International Conference on System Science and Engineering, ICSSE 2020 ; Conference date: 31-08-2020 Through 03-09-2020",

year = "2020",

month = aug,

doi = "10.1109/ICSSE50014.2020.9219301",

language = "English",

series = "2020 International Conference on System Science and Engineering, ICSSE 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2020 International Conference on System Science and Engineering, ICSSE 2020",

}

TY - GEN

T1 - Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks

AU - Chu, Tsai Ping

AU - Jhou, Jian Hua

AU - Leu, Yih Guang

PY - 2020/8

Y1 - 2020/8

N2 - The solar power variability is due to the variability of solar irradiance. Several factors are involved in the situation, such as cloud thickness and air pollution. In this paper, we attempt to find a novel way to predict the amount of solar irradiance. A image-based forecasting method is developed, and Long Short-Term Memory (LSTM) neural network is applied for data training. Daily solar irradiance and sky images are record by the record system, and uploaded to the MySQL database for storage. Feature values obtained by analyzing sky images are used as the input of neural network with solar irradiance. After some performance evaluation indicators were demonstrated, we found that the proposed method has good predictive performance with 5 to 60 minutes in present.

AB - The solar power variability is due to the variability of solar irradiance. Several factors are involved in the situation, such as cloud thickness and air pollution. In this paper, we attempt to find a novel way to predict the amount of solar irradiance. A image-based forecasting method is developed, and Long Short-Term Memory (LSTM) neural network is applied for data training. Daily solar irradiance and sky images are record by the record system, and uploaded to the MySQL database for storage. Feature values obtained by analyzing sky images are used as the input of neural network with solar irradiance. After some performance evaluation indicators were demonstrated, we found that the proposed method has good predictive performance with 5 to 60 minutes in present.

KW - Deep Learning

KW - Image Feature

KW - Recurrent Neural Network

KW - Solar Irradiance Forecasting

UR - http://www.scopus.com/inward/record.url?scp=85095569092&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85095569092&partnerID=8YFLogxK

U2 - 10.1109/ICSSE50014.2020.9219301

DO - 10.1109/ICSSE50014.2020.9219301

M3 - Conference contribution

AN - SCOPUS:85095569092

T3 - 2020 International Conference on System Science and Engineering, ICSSE 2020

BT - 2020 International Conference on System Science and Engineering, ICSSE 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 International Conference on System Science and Engineering, ICSSE 2020

Y2 - 31 August 2020 through 3 September 2020

ER -

Image-based Solar Irradiance Forecasting Using Recurrent Neural Networks

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this