Defining the best-fit machine learning classifier to early diagnose photovoltaic solar cells hot-spots

Mahmoud Dhimish

doi:10.1016/j.csite.2021.100980

Defining the best-fit machine learning classifier to early diagnose photovoltaic solar cells hot-spots

Mahmoud Dhimish

Electronic Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Photovoltaic (PV) hot-spots is a reliability problem in PV modules, where a cell or group of cells heats up significantly, dissipating rather than producing power, and resulting in a loss and further degradation for the PV modules’ performance. Therefore, in this article, we present the development of a novel machine learning-based (ML) tool to diagnose early-stage PV hot-spots. To achieve the best-fit ML structure, we compared four distinct machine learning classifiers, including decision tree (DT), support vector machine (SVM), K-nearest neighbour (KNN), and the discriminant classifiers (DC). Results confirm that the DC classifiers attain the best detection accuracy of 98%, while the least detection accuracy of 84% was observed for the decision tree. Furthermore, the examined four classifiers were also compared in terms of their performance using the confusion matrix and the receiver operating characteristics (ROC).

Original language	English
Article number	100980
Journal	Case Studies in Thermal Engineering
Volume	25
Early online date	7 Apr 2021
DOIs	https://doi.org/10.1016/j.csite.2021.100980
Publication status	Published - 1 Jun 2021

Bibliographical note

Access to Document

10.1016/j.csite.2021.100980Licence: CC BY

1-s2.0-S2214157X2100143X-main
© 2021 The Author(s)
Final published version, 4.43 MBLicence: CC BY

Cite this

@article{718cbaef8ed54145b1974189f6f6c64f,

title = "Defining the best-fit machine learning classifier to early diagnose photovoltaic solar cells hot-spots",

abstract = "Photovoltaic (PV) hot-spots is a reliability problem in PV modules, where a cell or group of cells heats up significantly, dissipating rather than producing power, and resulting in a loss and further degradation for the PV modules{\textquoteright} performance. Therefore, in this article, we present the development of a novel machine learning-based (ML) tool to diagnose early-stage PV hot-spots. To achieve the best-fit ML structure, we compared four distinct machine learning classifiers, including decision tree (DT), support vector machine (SVM), K-nearest neighbour (KNN), and the discriminant classifiers (DC). Results confirm that the DC classifiers attain the best detection accuracy of 98%, while the least detection accuracy of 84% was observed for the decision tree. Furthermore, the examined four classifiers were also compared in terms of their performance using the confusion matrix and the receiver operating characteristics (ROC).",

author = "Mahmoud Dhimish",

note = "{\textcopyright} 2021 The Author(s)",

year = "2021",

month = jun,

day = "1",

doi = "10.1016/j.csite.2021.100980",

language = "English",

volume = "25",

journal = "Case Studies in Thermal Engineering",

issn = "2214-157X",

publisher = "Elsevier",

}

TY - JOUR

T1 - Defining the best-fit machine learning classifier to early diagnose photovoltaic solar cells hot-spots

AU - Dhimish, Mahmoud

PY - 2021/6/1

Y1 - 2021/6/1

N2 - Photovoltaic (PV) hot-spots is a reliability problem in PV modules, where a cell or group of cells heats up significantly, dissipating rather than producing power, and resulting in a loss and further degradation for the PV modules’ performance. Therefore, in this article, we present the development of a novel machine learning-based (ML) tool to diagnose early-stage PV hot-spots. To achieve the best-fit ML structure, we compared four distinct machine learning classifiers, including decision tree (DT), support vector machine (SVM), K-nearest neighbour (KNN), and the discriminant classifiers (DC). Results confirm that the DC classifiers attain the best detection accuracy of 98%, while the least detection accuracy of 84% was observed for the decision tree. Furthermore, the examined four classifiers were also compared in terms of their performance using the confusion matrix and the receiver operating characteristics (ROC).

AB - Photovoltaic (PV) hot-spots is a reliability problem in PV modules, where a cell or group of cells heats up significantly, dissipating rather than producing power, and resulting in a loss and further degradation for the PV modules’ performance. Therefore, in this article, we present the development of a novel machine learning-based (ML) tool to diagnose early-stage PV hot-spots. To achieve the best-fit ML structure, we compared four distinct machine learning classifiers, including decision tree (DT), support vector machine (SVM), K-nearest neighbour (KNN), and the discriminant classifiers (DC). Results confirm that the DC classifiers attain the best detection accuracy of 98%, while the least detection accuracy of 84% was observed for the decision tree. Furthermore, the examined four classifiers were also compared in terms of their performance using the confusion matrix and the receiver operating characteristics (ROC).

UR - https://www.sciencedirect.com/science/article/pii/S2214157X2100143X?via%3Dihub

U2 - 10.1016/j.csite.2021.100980

DO - 10.1016/j.csite.2021.100980

M3 - Article

SN - 2214-157X

VL - 25

JO - Case Studies in Thermal Engineering

JF - Case Studies in Thermal Engineering

M1 - 100980

ER -

Defining the best-fit machine learning classifier to early diagnose photovoltaic solar cells hot-spots

Abstract

Bibliographical note

Access to Document

Other files and links

Cite this