An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance

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

3 Citations (Scopus)

Abstract

In this paper we propose a hybridized approach for finding high quality artificial neural network (ANN) for calculating hourly estimates of solar irradiance. These properties are essential for performance analysis of solar based energy generation. To be more precise the hourly global horizontal irradiance (GHI), direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI) are estimated based on ANNs which are trained using satellite and ground measurement data. In the proposed method we explore the effect of combining the measured data with properties derived from the standard physical models. The performance of the method is improved by using a genetic algorithm in two ways. First by selecting the parameters that are used for training the ANN. Secondly by adapting the size of the hidden layer of the ANN based on the number of selected input parameters. The adaptive size based approach proves to be especially suitable for ANN ensembles. In our computational experiments we evaluate the effectiveness of the proposed method on feedforward neural network. The results show that the adaptability of the ANN manages to notably improve the performance when compared to the standard approach using a fixed size of the hidden layer.

Original languageEnglish
Title of host publication2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1962-1967
Number of pages6
Volume2017-May
ISBN (Electronic)9781509061815
DOIs
Publication statusPublished - 30 Jun 2017
Event2017 International Joint Conference on Neural Networks, IJCNN 2017 - Anchorage, United States
Duration: 14 May 201719 May 2017

Other

Other2017 International Joint Conference on Neural Networks, IJCNN 2017
CountryUnited States
CityAnchorage
Period14/5/1719/5/17

Fingerprint

Neural networks
Feedforward neural networks
Genetic algorithms
Satellites
Experiments

Keywords

  • Artificial neural network
  • Ensemble model
  • Evolutionary artificial neural network
  • Global solar radiation

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Jovanovic, R., Martin Pomare Pomares, L., Mohieldeen, Y., Astudillo, D., & Bachour, D. (2017). An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance. In 2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings (Vol. 2017-May, pp. 1962-1967). [7966091] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2017.7966091

An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance. / Jovanovic, Raka; Martin Pomare Pomares, Luis; Mohieldeen, Yasir; Astudillo, Daniel; Bachour, Dunia.

2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings. Vol. 2017-May Institute of Electrical and Electronics Engineers Inc., 2017. p. 1962-1967 7966091.

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

Jovanovic, R, Martin Pomare Pomares, L, Mohieldeen, Y, Astudillo, D & Bachour, D 2017, An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance. in 2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings. vol. 2017-May, 7966091, Institute of Electrical and Electronics Engineers Inc., pp. 1962-1967, 2017 International Joint Conference on Neural Networks, IJCNN 2017, Anchorage, United States, 14/5/17. https://doi.org/10.1109/IJCNN.2017.7966091
Jovanovic R, Martin Pomare Pomares L, Mohieldeen Y, Astudillo D, Bachour D. An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance. In 2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings. Vol. 2017-May. Institute of Electrical and Electronics Engineers Inc. 2017. p. 1962-1967. 7966091 https://doi.org/10.1109/IJCNN.2017.7966091
Jovanovic, Raka ; Martin Pomare Pomares, Luis ; Mohieldeen, Yasir ; Astudillo, Daniel ; Bachour, Dunia. / An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance. 2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings. Vol. 2017-May Institute of Electrical and Electronics Engineers Inc., 2017. pp. 1962-1967
@inproceedings{4ac0b472ea3a4b0c829284d147499101,
title = "An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance",
abstract = "In this paper we propose a hybridized approach for finding high quality artificial neural network (ANN) for calculating hourly estimates of solar irradiance. These properties are essential for performance analysis of solar based energy generation. To be more precise the hourly global horizontal irradiance (GHI), direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI) are estimated based on ANNs which are trained using satellite and ground measurement data. In the proposed method we explore the effect of combining the measured data with properties derived from the standard physical models. The performance of the method is improved by using a genetic algorithm in two ways. First by selecting the parameters that are used for training the ANN. Secondly by adapting the size of the hidden layer of the ANN based on the number of selected input parameters. The adaptive size based approach proves to be especially suitable for ANN ensembles. In our computational experiments we evaluate the effectiveness of the proposed method on feedforward neural network. The results show that the adaptability of the ANN manages to notably improve the performance when compared to the standard approach using a fixed size of the hidden layer.",
keywords = "Artificial neural network, Ensemble model, Evolutionary artificial neural network, Global solar radiation",
author = "Raka Jovanovic and {Martin Pomare Pomares}, Luis and Yasir Mohieldeen and Daniel Astudillo and Dunia Bachour",
year = "2017",
month = "6",
day = "30",
doi = "10.1109/IJCNN.2017.7966091",
language = "English",
volume = "2017-May",
pages = "1962--1967",
booktitle = "2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - An evolutionary method for creating ensembles with adaptive size neural networks for predicting hourly solar irradiance

AU - Jovanovic, Raka

AU - Martin Pomare Pomares, Luis

AU - Mohieldeen, Yasir

AU - Astudillo, Daniel

AU - Bachour, Dunia

PY - 2017/6/30

Y1 - 2017/6/30

N2 - In this paper we propose a hybridized approach for finding high quality artificial neural network (ANN) for calculating hourly estimates of solar irradiance. These properties are essential for performance analysis of solar based energy generation. To be more precise the hourly global horizontal irradiance (GHI), direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI) are estimated based on ANNs which are trained using satellite and ground measurement data. In the proposed method we explore the effect of combining the measured data with properties derived from the standard physical models. The performance of the method is improved by using a genetic algorithm in two ways. First by selecting the parameters that are used for training the ANN. Secondly by adapting the size of the hidden layer of the ANN based on the number of selected input parameters. The adaptive size based approach proves to be especially suitable for ANN ensembles. In our computational experiments we evaluate the effectiveness of the proposed method on feedforward neural network. The results show that the adaptability of the ANN manages to notably improve the performance when compared to the standard approach using a fixed size of the hidden layer.

AB - In this paper we propose a hybridized approach for finding high quality artificial neural network (ANN) for calculating hourly estimates of solar irradiance. These properties are essential for performance analysis of solar based energy generation. To be more precise the hourly global horizontal irradiance (GHI), direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI) are estimated based on ANNs which are trained using satellite and ground measurement data. In the proposed method we explore the effect of combining the measured data with properties derived from the standard physical models. The performance of the method is improved by using a genetic algorithm in two ways. First by selecting the parameters that are used for training the ANN. Secondly by adapting the size of the hidden layer of the ANN based on the number of selected input parameters. The adaptive size based approach proves to be especially suitable for ANN ensembles. In our computational experiments we evaluate the effectiveness of the proposed method on feedforward neural network. The results show that the adaptability of the ANN manages to notably improve the performance when compared to the standard approach using a fixed size of the hidden layer.

KW - Artificial neural network

KW - Ensemble model

KW - Evolutionary artificial neural network

KW - Global solar radiation

UR - http://www.scopus.com/inward/record.url?scp=85031004782&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031004782&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2017.7966091

DO - 10.1109/IJCNN.2017.7966091

M3 - Conference contribution

VL - 2017-May

SP - 1962

EP - 1967

BT - 2017 International Joint Conference on Neural Networks, IJCNN 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -