Uncertainties of PV Module - Long-Term Outdoor Testing

Abstract

Usually, the measurement of the electrical performance of photovoltaic (PV) modules at standard test conditions (STC: 25°C junction temperature, 1000 Wm-2 plane of module irradiance and AM1.5G spectrum) is carried out indoors with a solar simulator or outdoors on a sun-tracker during a short-term test. Whitfield and Osterwald proposed a calculation procedure for the uncertainty of PV module performance measurement for the case of short-term outdoor measurement. The purpose of the present work is to evaluate this calculation procedure in the case of IV- parameters extraction from data acquired during long-term outdoor testing of PV modules mounted on a fixed stand, in Tallinn, Estonia. In such a northern location, the data distributions present a large amount of points at condition of standard irradiance and temperature (SIT: 1000 Wm-2 and 25°C). An uncertainty analysis dedicated to this situation is proposed. It includes the calculation procedure proposed by Whitfield and Osterwald which allows for a straightforward calculation of the uncertainties in parameters obtained by linear regression methods when the experimental data points have uncertainties (error bars) in both coordinates. The procedure of Whitfield and Osterwald provides consistent uncertainties compared with linear regression methods taking into account error bars in the y-direction only.