The characterisation of crystalline silicon photovoltaic devices using the manufacturer supplied data. (July 2016)

• Record Type: Journal Article
• Authors: Ghani, F.; Rosengarten, G.; Duke, M.
• Is Part Of: Solar energy. Volume 132(2016)
• Page Start: 15
• Page End: 24
• Publication Date: 2016-07
• Subjects: Solar cell -- Photovoltaic -- Characterisation -- Parameter values -- Modelling; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2016.03.008
• Languages: English
• ISSNs: 0038-092X

Abstract

Highlights: A method is presented to determine the characterisation parameters of a photovoltaic cell. The proposed method requires only data typically provided by the manufacturer. Method has been experimentally validated and both qualitatively and quantitatively assessed. Provides a method to assist with the accurate modelling of a photovoltaic devices. Root mean square error less than 1% for conditions investigated. Abstract: Currently, the single diode, five parameter model is extensively used to mathematically model the electrical behaviour of a photovoltaic device. The accuracy of this model is, however, strongly dependent on a number factors but particularly the quality of the five parameters used to characterise the devices unique performance behaviour. These are; the series and shunt resistances, the photogenerated and reverse saturation currents, and the diode factor. In this study, a new method is presented to determine four of the five parameter values using the data typically provided by the manufacturer. The proposed method is based on calculating the values for the series and shunt resistances and the photogenerated and reverse saturation currents by solving the underlying non-linear equations at short-circuit, maximum-power-point, and open-circuit conditions numerically. The proposed algorithm was experimentally validated against a crystalline silicone type solar cell. The characterisation parameter values were calculated using the proposed algorithm in Matlab, and then used to generate the simulated I – V data. When the experimental data was compared against the generated simulated data, it was found that the simulated data closely matched the experimental data at the three chosen points (short circuit current, maximum power point, and open circuit voltage), with a minor disparity emerging outside these locations. This disparity was quantified using a root mean square error (RMSE) approach which revealed a maximum RMSE value of 0.0072 when assuming a diode factor ( n ) of one. By incrementally adjusting the diode ideality factor from n = 1 to 1.5, the RMSE was reduced further to 0.0008.