Modeling of photovoltaic array using random forests technique

Ibrahim A. Ibrahim, Azah Mohamed, Tamer Khatib

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

3 Citations (Scopus)

Abstract

This paper presents a novel technique for modeling of photovoltaic (PV) array using random forests (RFs). Metrological variables such as solar radiation and ambient temperature as well as actual output current of a 3 kWp PV grid-connected system installed at Universiti Kebangsaan Malaysia have been utilized. These data are used to train and validate the proposed RFs model. Three statistical error values, namely, root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE), are used to evaluate the developed model. The results show that the proposed RFs model accurately predicts the output current of the PV system. The RMSE, MAPE, and MBE values of the RFs model are 2.7482%, 8.7151%, and -2.5772%, respectively.

Original languageEnglish
Title of host publication2015 IEEE Conference on Energy Conversion, CENCON 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages390-393
Number of pages4
ISBN (Print)9781479985982
DOIs
Publication statusPublished - 16 Feb 2016
Event2nd IEEE Conference on Energy Conversion, CENCON 2015 - Johor Bahru, Malaysia
Duration: 19 Oct 201520 Oct 2015

Other

Other2nd IEEE Conference on Energy Conversion, CENCON 2015
CountryMalaysia
CityJohor Bahru
Period19/10/1520/10/15

Fingerprint

Mean square error
Solar radiation
Temperature

Keywords

  • modeling of PV systems
  • performance evaluation
  • random forests

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Cite this

Ibrahim, I. A., Mohamed, A., & Khatib, T. (2016). Modeling of photovoltaic array using random forests technique. In 2015 IEEE Conference on Energy Conversion, CENCON 2015 (pp. 390-393). [7409575] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CENCON.2015.7409575

Modeling of photovoltaic array using random forests technique. / Ibrahim, Ibrahim A.; Mohamed, Azah; Khatib, Tamer.

2015 IEEE Conference on Energy Conversion, CENCON 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 390-393 7409575.

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

Ibrahim, IA, Mohamed, A & Khatib, T 2016, Modeling of photovoltaic array using random forests technique. in 2015 IEEE Conference on Energy Conversion, CENCON 2015., 7409575, Institute of Electrical and Electronics Engineers Inc., pp. 390-393, 2nd IEEE Conference on Energy Conversion, CENCON 2015, Johor Bahru, Malaysia, 19/10/15. https://doi.org/10.1109/CENCON.2015.7409575
Ibrahim IA, Mohamed A, Khatib T. Modeling of photovoltaic array using random forests technique. In 2015 IEEE Conference on Energy Conversion, CENCON 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 390-393. 7409575 https://doi.org/10.1109/CENCON.2015.7409575
Ibrahim, Ibrahim A. ; Mohamed, Azah ; Khatib, Tamer. / Modeling of photovoltaic array using random forests technique. 2015 IEEE Conference on Energy Conversion, CENCON 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 390-393
@inproceedings{b38d0719c6fc47c68d188695c48def28,
title = "Modeling of photovoltaic array using random forests technique",
abstract = "This paper presents a novel technique for modeling of photovoltaic (PV) array using random forests (RFs). Metrological variables such as solar radiation and ambient temperature as well as actual output current of a 3 kWp PV grid-connected system installed at Universiti Kebangsaan Malaysia have been utilized. These data are used to train and validate the proposed RFs model. Three statistical error values, namely, root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE), are used to evaluate the developed model. The results show that the proposed RFs model accurately predicts the output current of the PV system. The RMSE, MAPE, and MBE values of the RFs model are 2.7482{\%}, 8.7151{\%}, and -2.5772{\%}, respectively.",
keywords = "modeling of PV systems, performance evaluation, random forests",
author = "Ibrahim, {Ibrahim A.} and Azah Mohamed and Tamer Khatib",
year = "2016",
month = "2",
day = "16",
doi = "10.1109/CENCON.2015.7409575",
language = "English",
isbn = "9781479985982",
pages = "390--393",
booktitle = "2015 IEEE Conference on Energy Conversion, CENCON 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Modeling of photovoltaic array using random forests technique

AU - Ibrahim, Ibrahim A.

AU - Mohamed, Azah

AU - Khatib, Tamer

PY - 2016/2/16

Y1 - 2016/2/16

N2 - This paper presents a novel technique for modeling of photovoltaic (PV) array using random forests (RFs). Metrological variables such as solar radiation and ambient temperature as well as actual output current of a 3 kWp PV grid-connected system installed at Universiti Kebangsaan Malaysia have been utilized. These data are used to train and validate the proposed RFs model. Three statistical error values, namely, root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE), are used to evaluate the developed model. The results show that the proposed RFs model accurately predicts the output current of the PV system. The RMSE, MAPE, and MBE values of the RFs model are 2.7482%, 8.7151%, and -2.5772%, respectively.

AB - This paper presents a novel technique for modeling of photovoltaic (PV) array using random forests (RFs). Metrological variables such as solar radiation and ambient temperature as well as actual output current of a 3 kWp PV grid-connected system installed at Universiti Kebangsaan Malaysia have been utilized. These data are used to train and validate the proposed RFs model. Three statistical error values, namely, root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE), are used to evaluate the developed model. The results show that the proposed RFs model accurately predicts the output current of the PV system. The RMSE, MAPE, and MBE values of the RFs model are 2.7482%, 8.7151%, and -2.5772%, respectively.

KW - modeling of PV systems

KW - performance evaluation

KW - random forests

UR - http://www.scopus.com/inward/record.url?scp=84964317410&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964317410&partnerID=8YFLogxK

U2 - 10.1109/CENCON.2015.7409575

DO - 10.1109/CENCON.2015.7409575

M3 - Conference contribution

SN - 9781479985982

SP - 390

EP - 393

BT - 2015 IEEE Conference on Energy Conversion, CENCON 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -