Open AccessTracking the performance of photovoltaic systems: a tool for minimising the risk of malfunctions and deterioration
Author(s) -
Spiliotis Evangelos,
Legaki Nikoletta Zampeta,
Assimakopoulos Vassilios,
Doukas Haris,
El Moursi Mohamed Shawky
Publication year - 2018
Publication title -
iet renewable power generation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.005
H-Index - 76
eISSN - 1752-1424
pISSN - 1752-1416
DOI - 10.1049/iet-rpg.2017.0596
Subject(s) - photovoltaic system , component (thermodynamics) , computer science , production (economics) , reliability engineering , relation (database) , artificial neural network , tracking (education) , energy (signal processing) , risk analysis (engineering) , operations research , engineering , artificial intelligence , data mining , business , statistics , physics , mathematics , electrical engineering , economics , macroeconomics , thermodynamics , psychology , pedagogy
The environmental and economic impact of photovoltaic (PV) systems is continuously growing, serving as an effective alternative energy source. Yet, failures and underperformance, e.g. due to soiling and deterioration, can significantly affect PV production and shrink the capacity available. This becomes an important issue, especially when the plant is not easily accessible for manual checking. Typical monitoring tools can help energy managers to deal with such issues. However, their diagnostics might be misleading as reduced performance could also be caused by low radiation and other relative factors, which are difficult to identify given the non‐linear and stochastic relation of energy production and weather variables. In addition, accurate component‐based models that use local weather measurements as inputs are not always applicable. In this regard, a methodological approach for tracking the performance of PV systems is proposed, which uses an artificial neural network, trained using reported system data and irradiation predictions. Possible abnormalities are identified through the model and alerts are generated to proceed with maintenance actions. The approach is implemented into a decision support system for smart cities, demonstrating encouraging results.