In the last years, several numerical techniques were proposed in literature for the identification of the one diode model of photovoltaic (PV) panels from manufacturers’ data sheets or experimental data. In this paper we present a fast and accurate method for obtaining the five parameters of the one diode model by starting from the experimental I–V curve of the PV panel. In particular we exploit the adoption of the reduced forms of the original five-parameter model. These reduced forms take advantage of important mathematical considerations through which the five parameters of the model are splitted in independent and dependent unknowns, in order to reduce the dimensions of the search space. Thus, the fitting of experimental data can be performed by using a restricted number of unknowns (namely two) and this provides strong benefits in terms of convergence, computational costs and execution times. In this way, the approach allows to characterize a PV panel from its measured I–V curve with an accuracy never obtained before in literature, employing few steps and execution times of few milliseconds on a simple notebook computer. Suitable validations on two important case studies are presented for comparing our procedure with the most recent and effective techniques proposed in literature.
High performing extraction procedure for the one-diode model of a photovoltaic panel from experimental I–V curves by using reduced forms
LAUDANI, ANTONINO
;
2014-01-01
Abstract
In the last years, several numerical techniques were proposed in literature for the identification of the one diode model of photovoltaic (PV) panels from manufacturers’ data sheets or experimental data. In this paper we present a fast and accurate method for obtaining the five parameters of the one diode model by starting from the experimental I–V curve of the PV panel. In particular we exploit the adoption of the reduced forms of the original five-parameter model. These reduced forms take advantage of important mathematical considerations through which the five parameters of the model are splitted in independent and dependent unknowns, in order to reduce the dimensions of the search space. Thus, the fitting of experimental data can be performed by using a restricted number of unknowns (namely two) and this provides strong benefits in terms of convergence, computational costs and execution times. In this way, the approach allows to characterize a PV panel from its measured I–V curve with an accuracy never obtained before in literature, employing few steps and execution times of few milliseconds on a simple notebook computer. Suitable validations on two important case studies are presented for comparing our procedure with the most recent and effective techniques proposed in literature.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.