Mathematical and neural network modeling for predicting and analyzing of nanofluid-nano PCM photovoltaic thermal systems performance

Ali H.A. Al-Waeli, Hussein A. Kazem, Jabar H. Yousif, Miqdam T. Chaichan, Kamaruzzaman Sopian

Research output: Contribution to journalArticle

Abstract

This paper aims to enhance the power production performance of the PV/T based on three cooling models using nanofluid, SiC-water and nano-PCM. The effect of solar irradiance and ambient temperature on the power productivity performance has been investigated based on sensitivity analysis that shows that electrical current and solar irradiance has more impact on power prediction than voltage and temperature. Also, three mathematical linear prediction models were developed and compared with the prediction of ANN models, and the results were verified and fit the experimental results. The comparison with published literature is made based on three common evaluation criteria that include the coefficient of determination R2, MSE and RSME. The proposed predicting models attained an excellent R2 result of 0.99 and MSE value of 0.006 and RSME of 0.009 for both P-M1 and P-M2 models. Besides, the P-M3 obtained an MSE value of 0.022. Finally, the proposed linear prediction models help to reduce the error in furcating future results and determine the best conditions for any solar system in an easy and fast way.

Original languageEnglish
Pages (from-to)963-980
Number of pages18
JournalRenewable Energy
Volume145
DOIs
Publication statusPublished - 1 Jan 2020

Fingerprint

Pulse code modulation
Neural networks
Solar system
Sensitivity analysis
Hot Temperature
Productivity
Cooling
Temperature
Electric potential
Water

Keywords

  • Nano-PCM
  • Nano-sic
  • Nanofluid
  • Neural network
  • PV/T
  • Statistical analysis

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment

Cite this

Mathematical and neural network modeling for predicting and analyzing of nanofluid-nano PCM photovoltaic thermal systems performance. / Al-Waeli, Ali H.A.; Kazem, Hussein A.; Yousif, Jabar H.; Chaichan, Miqdam T.; Sopian, Kamaruzzaman.

In: Renewable Energy, Vol. 145, 01.01.2020, p. 963-980.

Research output: Contribution to journalArticle

Al-Waeli, Ali H.A. ; Kazem, Hussein A. ; Yousif, Jabar H. ; Chaichan, Miqdam T. ; Sopian, Kamaruzzaman. / Mathematical and neural network modeling for predicting and analyzing of nanofluid-nano PCM photovoltaic thermal systems performance. In: Renewable Energy. 2020 ; Vol. 145. pp. 963-980.
@article{6b4a9169b0134954ba943dda38981ce4,
title = "Mathematical and neural network modeling for predicting and analyzing of nanofluid-nano PCM photovoltaic thermal systems performance",
abstract = "This paper aims to enhance the power production performance of the PV/T based on three cooling models using nanofluid, SiC-water and nano-PCM. The effect of solar irradiance and ambient temperature on the power productivity performance has been investigated based on sensitivity analysis that shows that electrical current and solar irradiance has more impact on power prediction than voltage and temperature. Also, three mathematical linear prediction models were developed and compared with the prediction of ANN models, and the results were verified and fit the experimental results. The comparison with published literature is made based on three common evaluation criteria that include the coefficient of determination R2, MSE and RSME. The proposed predicting models attained an excellent R2 result of 0.99 and MSE value of 0.006 and RSME of 0.009 for both P-M1 and P-M2 models. Besides, the P-M3 obtained an MSE value of 0.022. Finally, the proposed linear prediction models help to reduce the error in furcating future results and determine the best conditions for any solar system in an easy and fast way.",
keywords = "Nano-PCM, Nano-sic, Nanofluid, Neural network, PV/T, Statistical analysis",
author = "Al-Waeli, {Ali H.A.} and Kazem, {Hussein A.} and Yousif, {Jabar H.} and Chaichan, {Miqdam T.} and Kamaruzzaman Sopian",
year = "2020",
month = "1",
day = "1",
doi = "10.1016/j.renene.2019.06.099",
language = "English",
volume = "145",
pages = "963--980",
journal = "Renewable Energy",
issn = "0960-1481",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Mathematical and neural network modeling for predicting and analyzing of nanofluid-nano PCM photovoltaic thermal systems performance

AU - Al-Waeli, Ali H.A.

AU - Kazem, Hussein A.

AU - Yousif, Jabar H.

AU - Chaichan, Miqdam T.

AU - Sopian, Kamaruzzaman

PY - 2020/1/1

Y1 - 2020/1/1

N2 - This paper aims to enhance the power production performance of the PV/T based on three cooling models using nanofluid, SiC-water and nano-PCM. The effect of solar irradiance and ambient temperature on the power productivity performance has been investigated based on sensitivity analysis that shows that electrical current and solar irradiance has more impact on power prediction than voltage and temperature. Also, three mathematical linear prediction models were developed and compared with the prediction of ANN models, and the results were verified and fit the experimental results. The comparison with published literature is made based on three common evaluation criteria that include the coefficient of determination R2, MSE and RSME. The proposed predicting models attained an excellent R2 result of 0.99 and MSE value of 0.006 and RSME of 0.009 for both P-M1 and P-M2 models. Besides, the P-M3 obtained an MSE value of 0.022. Finally, the proposed linear prediction models help to reduce the error in furcating future results and determine the best conditions for any solar system in an easy and fast way.

AB - This paper aims to enhance the power production performance of the PV/T based on three cooling models using nanofluid, SiC-water and nano-PCM. The effect of solar irradiance and ambient temperature on the power productivity performance has been investigated based on sensitivity analysis that shows that electrical current and solar irradiance has more impact on power prediction than voltage and temperature. Also, three mathematical linear prediction models were developed and compared with the prediction of ANN models, and the results were verified and fit the experimental results. The comparison with published literature is made based on three common evaluation criteria that include the coefficient of determination R2, MSE and RSME. The proposed predicting models attained an excellent R2 result of 0.99 and MSE value of 0.006 and RSME of 0.009 for both P-M1 and P-M2 models. Besides, the P-M3 obtained an MSE value of 0.022. Finally, the proposed linear prediction models help to reduce the error in furcating future results and determine the best conditions for any solar system in an easy and fast way.

KW - Nano-PCM

KW - Nano-sic

KW - Nanofluid

KW - Neural network

KW - PV/T

KW - Statistical analysis

UR - http://www.scopus.com/inward/record.url?scp=85067846055&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067846055&partnerID=8YFLogxK

U2 - 10.1016/j.renene.2019.06.099

DO - 10.1016/j.renene.2019.06.099

M3 - Article

VL - 145

SP - 963

EP - 980

JO - Renewable Energy

JF - Renewable Energy

SN - 0960-1481

ER -