Contents lists available at ScienceDirect Solar Energy journal homepage: www.elsevier.com/locate/solener Impacts of temperature and irradiance on polycrystalline silicon solar cells parameters D.M. Fébba ⁎ ,1 , R.M. Rubinger 1 , A.F. Oliveira 1 , E.C. Bortoni 1 Universidade Federal de Itajubá, Brazil ARTICLEINFO Keywords: Solar cells Solar cell parameters Temperature dependence Illumination intensity Diferential Evolution Polycrystalline silicon ABSTRACT The accurate knowledge of the solar cells parameters dependence on irradiance and temperature is of vital importancefortheperformanceassessmentofphotovoltaicmodulesanddevelopmentofnewdevices,andmany works have been published so far to understand the aforementioned dependence, but none employed a meta- heuristic technique. To understand the temperature and irradiance impacts on the single-diode parameters, sevenpolycrystallinesiliconsolarcellswerestudiedthroughacarefulexperimentalcharacterizationintherange of 600–1000W/m 2 and 25–55°C. To extract single-diode parameters, the Diferential Evolution optimization techniquewasemployed,resultinginverylowfttingerrorsbetweenexperimentalandsimulatedI-Vcurves.The results obtained showed that the shunt and series resistance were more afected by the increasing temperature, with an exponential decrease, than for increasing irradiance, with a linear increase and decrease for series and shunt resistance, respectively. The diode ideality factor showed no signifcant changes with increasing tem- perature and irradiance, while the diode saturation current showed an exponential dependence on increasing temperature,butnosignifcantchangeswithincreasingirradiance.Furthermore,itwasseenthatevenwithsame nominal features, polycrystalline silicon solar cells may present very diferent values and behaviors for the single-diode parameters. 1. Introduction The accurate modeling of solar cells is essential to understand and predict how photovoltaic devices operate under diferent temperature and irradiance conditions, considering that these devices generally operate in non-standard conditions (25°C and 1000W/m 2 )(Durisch et al., 1996). The most important parameters for the performance evaluation of a solar cell are extracted from the current-voltage (I-V) characteristics and, among the mathematical models available to de- scribe these I-V curves, the most widely used is the single-diode model, due to its simplicity and accurate results (Abbassi et al., 2018; Humada et al., 2016). The equation describing this model can be written as: = + + I I I exp qV IR nkT V IR R ( ) 1 , ph s s sh 0 (1) where I V qkTI I nR , , , , , , , , ph s 0 and R sh are the current (A) and voltage (V) on the solar cell terminals, the electronic charge (1.602 176 62×10 −19 C), Boltzmann constant (1.38064852×10 −23 JK −1 ), temperature (K), photocurrent (A), diode saturation current (A), diode ideality factor, series and shunt resistance ( ), respectively. Although this model is simpler when compared to others that take into account resistive losses and diode non-idealities, the extraction of the fve electrical parameters of interest (I I nR R , , , , ph s sh 0 ) for a given temperature and illumination is a challenging task, since Eq. (1) is a transcendental and non-linear equation. The problem of extracting the fve single-diode parameters from the I-V characteristics may be classifed into three main categories, ac- cording to the approach employed: analytical, numerical and soft computing algorithms (Lin et al., 2017). Analytical techniques require the knowledge of some key points of the I-V characteristics and slopes at short-circuit and open circuit regions (Ishaque et al., 2012). In this way, these methods rely on the choice of some regions of the I-V characteristics to perform curve fttings, and if these regions are not properly selected, signifcant errors may occur (Bühler et al., 2014). Numerical techniques can provide better results than the analytical onessinceallI-Vpointsareutilized,butitsaccuracydependonthetype of the ftting algorithm and on the chosen initial points (Chellaswamy and Ramesh, 2016; Ishaque et al., 2012). The metaheuristic techniques, https://doi.org/10.1016/j.solener.2018.09.051 Received 21 June 2018; Received in revised form 5 September 2018; Accepted 17 September 2018 ⁎ Corresponding author. 1 Address: Av. BPS, 1303, Pinheirinho, Itajubá, MG 37500-903, Brazil. E-mail addresses: davifebba@uol.com.br (D.M. Fébba), rero@unifei.edu.br (R.M. Rubinger), adhimarfavio@unifei.edu.br (A.F. Oliveira), bortoni@unifei.edu.br (E.C. Bortoni). Solar Energy 174 (2018) 628–639 0038-092X/ © 2018 Elsevier Ltd. All rights reserved. T