ABSTRACT
The deployment of photovoltaic technology in global electricity generation has reduced the dependence on fossil fuels and the adverse effect of global warming climate changes. However, the variability of PV output due to its reliance on meteorological parameters hinders the total maximization of the technology. In this paper, the interdependence of meteorological parameters on accurate estimation of the specific yield of the PV system is examined using machine learning algorithms. The analysis uses 2 years of meteorological data in a 10-min interval. The meteorological parameters used as input are global solar radiation (GR), wind speed (WS), wind direction (WD), the standard deviation of wind direction (WSD), air temperature (AT), and relative humidity (RH), while the output variable is the PV specific yield (PVSY). This study uses a Python-based “all regressor” machine learning algorithm, to analyze their accuracy for the estimation analysis at different input feature space. The accuracy of the models and algorithms is analyzed using mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and coefficient of correlation (R2). The best performance in the two feature space was retrieved for AT, and GR with MSE and MAE values of 0.113 and 0.252, respectively. The best performance in the three feature space was retrieved for WSD, AT, and GR with MSE and MAE values of 0.109 and 0.255, respectively. The results show that the most accurate PVSY estimation was for the four input feature space with input variables of WS, AT, GR, and RH. The K-nearest neighbor (KNN) proved to be the most appropriate algorithm for this feature space with an RMSE and MAE score of 0.101 and 0.266, respectively. The study also shows that the most significant meteorological inputs in PV yield estimation are air temperature and global solar radiation. The study also noted that a large feature space adds uncertainties to the training process. Finally, the paper concludes that the sunshine-based parameters are key inputs for high-quality PV yield estimation regardless of the number of input combinations. It is also concluded that, in this study, the most accurate regression algorithms for estimation are the Least Absolute Shrinkage and Selection Operator (LASSO), KNN, Elastic Net (EN), Decision Tree regressor (DT), and Extra Tree regression (ET).
Abbreviations
PVSYPhotovoltaic specific yield
KNNK Nearest neighbor
LASSOLeast absolute shrinkage and selection operators
MAEMean absolute error
MSEMean square error
RMSERoot mean square error
PVPhotovoltaic
DTDecision tree
ENElastic Net
RANSACRandom sample consensus
LSTMLong short-term memory
ANNArtificialneural network
SVMSupport vector machine
WSWind speed
WDWind Direction
WSDStandard deviation of wind speed
ATAir temperature-
GRGlobal radiation
RHRelative humidity
ETExtra tree
CCACanonical correlation analysis
SVRSupport vector regression
Additional information
Notes on contributors
Humphrey Adun
Humphrey Adun received his master’s degree in Energy systems engineering from Cyprus International University (CIU), in 2017. He is currently pursuing a PhD degree in Energy Systems Engineering at Cyprus International University (CIU). He has worked as a research assistant, assistant lecturer, and lecturer between 2016 and 2021. His research interests include renewable energy integration, nanofluid application in solar thermal systems, and energy policy
Olusola Bamisile
Olusola Bamisile received his master’s degree in Energy systems engineering from Cyprus International University (CIU), in 2016. He is currently pursuing a PhD degree in Electronic Science and Technology at the University of Electronic Science and Technology of China (UESTC). He has worked as a research assistant, assistant lecturer, and lecturer between 2016 and 2018, with a great wealth of experience also in laboratory activities during this period. His research interests include renewable energy integration, comprehensive energy system modeling, energy planning, and machine learning application for solar energy production.
Mustapha Mukhtar
Dr. Mukhtar Mustapha is a senior lectuere with school of Economics and Management, Guangdong University of Petrochemical Technology Maoming. He earned his Ph.D from Universiti Utara Malaysia in 2019/2020 with specialty in Economics. his field of interest include among others renewable energy and enviromental economics. he has supervised more thn 30 students in undergraduate and Msc, he also has more than 10 publication to his credit.
Mustafa Dagbasi
Mustafa Dagbasi is one of the outstanding mechanical engineers in the Turkish Republic of North Cyprus. Having worked as the Vice Rector at Eastern Mediterranean University in time past, he is currently the Head of Mechanical Engineering and Energy Systems Engineering in Cyprus International University. His research focus is HVAC system applications and thermodynamics.
Doga Kavaz
Professor Doga Kavazis one of the most outstanding Nanotechnologist in Turkish Republic of North Cyprus. She has tremendous experience and research discoveries in nanotechnology.
Ariyo Oluwasanmi
Ariyo Oluwasanmi received his master’s degree in software engineering in 2017, and his Ph.D. degree in software engineering in June 2020 from the University of Electronic Science and Technology of China (UESTC). He is a senior researcher and a team lead with the UESTC Deep Learning Workshop and his research interests include the application of deep learning and reinforcement learning to computer vision and artificial general intelligence.