Comparative analysis of parameter extraction techniques for the electrical characterization of multi-junction CPV and m-Si technologies. (15th January 2018)

• Record Type: Journal Article
• Authors: Montes-Romero, Jesús; Almonacid, Florencia; Theristis, Marios; de la Casa, Juan; Georghiou, George E.; Fernández, Eduardo F.
• Is Part Of: Solar energy. Volume 160(2018)
• Page Start: 275
• Page End: 288
• Publication Date: 2018-01-15
• Subjects: Concentrating photovoltaics -- Monocrystalline silicon -- Single-diode model -- Outdoor characterisation; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2017.12.011
• Languages: English
• ISSNs: 0038-092X

Abstract

Highlights: The single exponential model is widely used to predict the IV curve of PV devices. Extraction methods are applied and compared for a m-Si and a concentrator PV module. The models are analysed with long-term data of the modules installed side-by-side. The methods are evaluated against the irradiance, module temperature and air mass. The impact of irradiance on the parameters is studied and fitting equations derived. Abstract: Modelling the current-voltage ( I-V ) characteristics of photovoltaic (PV) modules under real operating conditions is crucial for the better understanding of each technology. The most commonly used model for the electrical characterization of PV is the single-diode model where the five parameters are extracted through a variety of techniques. Although numerous extraction methods were developed for conventional PV technologies, the studies concerning the concentrating photovoltaic (CPV) technology are still limited. In this work, three analytical parameter extraction methods (Phang et al., Blas et al. and Khan et al.) are applied and compared for a multi-junction (MJ) CPV and a monocrystalline (m-Si) module based on long-term outdoor measurements in Jaén, Spain. The sensitivity of the models against the dominant parameters that influence the behaviour of PV and CPV (i.e. irradiance, module temperature and air mass) is also investigated. Furthermore, the effect of irradiance on the extracted parameters is discussed including the derivation of the corresponding fitting equations and errors. The results indicate that the most robust method is the one proposed by Phang et al. for the CPV module (normalised root mean square error, NRMSE, of 1.55%) and the one proposed by Blas et al. for the m-Si module ( NRMSE of 0.58%). The method of Khan et al. resulted the highest error values for every case ( NRMSE of 4.5% and 1.74% for CPV and m-Si respectively) while the Phang et al. method exhibited a similar error for both technologies. The main outcome of this work contributes to the optimum selection of parameter extraction techniques depending on the technology and the desired associated errors while the analysis of the dependence of the parameters on irradiance provides a better understanding of each technology's behaviour in the field.