Hotspot diagnosis for solar photovoltaic modules using a Naive Bayes classifier

Kamran Ali Khan Niazi, Wajahat Akhtar, Hassan A. Khan, Yongheng Yang*, Shahrukh Athar

*Kontaktforfatter

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

Monitoring and maintenance of photovoltaic (PV) modules are critical for a reliable and efficient operation. Hotspots in PV modules due to various defects and operational conditions may challenge the reliability, and in turn, the entire system. From the monitoring standpoint, hotspots should be detected and categorized for subsequent maintenance. In this paper, hotspots are detected, evaluated, and categorized uniquely by using a machine learning technique on thermal images of PV modules. To achieve so, the texture and histogram of gradient (HOG) features of thermal images of PV modules are used for classification. The categorized hotspots are detected by training the machine learning algorithm, i.e., a Naive Bayes (nBayes) classifier. Experimental results are performed on a 42.24-kWp PV system, which demonstrates that a mean recognition rate of around 94.1% is achieved for the set of 375 samples.
OriginalsprogEngelsk
TidsskriftSolar Energy
Vol/bind190
Sider (fra-til)34-43
Antal sider10
ISSN0038-092X
DOI
StatusUdgivet - sep. 2019

Fingerprint

Learning systems
Classifiers
Monitoring
Learning algorithms
Textures
Defects
Hot Temperature

Citer dette

Niazi, Kamran Ali Khan ; Akhtar, Wajahat ; Khan, Hassan A. ; Yang, Yongheng ; Athar, Shahrukh. / Hotspot diagnosis for solar photovoltaic modules using a Naive Bayes classifier. I: Solar Energy. 2019 ; Bind 190. s. 34-43.
@article{f36afb8bd416417fb1580efb5166cb80,
title = "Hotspot diagnosis for solar photovoltaic modules using a Naive Bayes classifier",
abstract = "Monitoring and maintenance of photovoltaic (PV) modules are critical for a reliable and efficient operation. Hotspots in PV modules due to various defects and operational conditions may challenge the reliability, and in turn, the entire system. From the monitoring standpoint, hotspots should be detected and categorized for subsequent maintenance. In this paper, hotspots are detected, evaluated, and categorized uniquely by using a machine learning technique on thermal images of PV modules. To achieve so, the texture and histogram of gradient (HOG) features of thermal images of PV modules are used for classification. The categorized hotspots are detected by training the machine learning algorithm, i.e., a Naive Bayes (nBayes) classifier. Experimental results are performed on a 42.24-kWp PV system, which demonstrates that a mean recognition rate of around 94.1{\%} is achieved for the set of 375 samples.",
keywords = "Photovoltaic (PV) modules, Monitoring, Hotspots, Texture and histogram of gradient (HOG) features, Thermographic assessment, Thermal images, Naive Bayes classifier, Machine learning",
author = "Niazi, {Kamran Ali Khan} and Wajahat Akhtar and Khan, {Hassan A.} and Yongheng Yang and Shahrukh Athar",
year = "2019",
month = "9",
doi = "10.1016/j.solener.2019.07.063",
language = "English",
volume = "190",
pages = "34--43",
journal = "Solar Energy",
issn = "0038-092X",
publisher = "Pergamon Press",

}

Hotspot diagnosis for solar photovoltaic modules using a Naive Bayes classifier. / Niazi, Kamran Ali Khan; Akhtar, Wajahat ; Khan, Hassan A.; Yang, Yongheng; Athar, Shahrukh.

I: Solar Energy, Bind 190, 09.2019, s. 34-43.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Hotspot diagnosis for solar photovoltaic modules using a Naive Bayes classifier

AU - Niazi, Kamran Ali Khan

AU - Akhtar, Wajahat

AU - Khan, Hassan A.

AU - Yang, Yongheng

AU - Athar, Shahrukh

PY - 2019/9

Y1 - 2019/9

N2 - Monitoring and maintenance of photovoltaic (PV) modules are critical for a reliable and efficient operation. Hotspots in PV modules due to various defects and operational conditions may challenge the reliability, and in turn, the entire system. From the monitoring standpoint, hotspots should be detected and categorized for subsequent maintenance. In this paper, hotspots are detected, evaluated, and categorized uniquely by using a machine learning technique on thermal images of PV modules. To achieve so, the texture and histogram of gradient (HOG) features of thermal images of PV modules are used for classification. The categorized hotspots are detected by training the machine learning algorithm, i.e., a Naive Bayes (nBayes) classifier. Experimental results are performed on a 42.24-kWp PV system, which demonstrates that a mean recognition rate of around 94.1% is achieved for the set of 375 samples.

AB - Monitoring and maintenance of photovoltaic (PV) modules are critical for a reliable and efficient operation. Hotspots in PV modules due to various defects and operational conditions may challenge the reliability, and in turn, the entire system. From the monitoring standpoint, hotspots should be detected and categorized for subsequent maintenance. In this paper, hotspots are detected, evaluated, and categorized uniquely by using a machine learning technique on thermal images of PV modules. To achieve so, the texture and histogram of gradient (HOG) features of thermal images of PV modules are used for classification. The categorized hotspots are detected by training the machine learning algorithm, i.e., a Naive Bayes (nBayes) classifier. Experimental results are performed on a 42.24-kWp PV system, which demonstrates that a mean recognition rate of around 94.1% is achieved for the set of 375 samples.

KW - Photovoltaic (PV) modules

KW - Monitoring

KW - Hotspots

KW - Texture and histogram of gradient (HOG) features

KW - Thermographic assessment

KW - Thermal images

KW - Naive Bayes classifier

KW - Machine learning

U2 - 10.1016/j.solener.2019.07.063

DO - 10.1016/j.solener.2019.07.063

M3 - Journal article

VL - 190

SP - 34

EP - 43

JO - Solar Energy

JF - Solar Energy

SN - 0038-092X

ER -