Novel method for the improvement in the evaluation of outdoor performance loss rate in different PV technologies and comparison with two other methods

• We present a new method to evaluate performance loss rate based on STC-corrected array generated power, Pmax,STCPmax,STC.
• The inclusion of spectral correction optimizes PmaxPmax method accuracy.
• We compare the associated uncertainty with other well-known methods (PR and PVUSA).
• Pmax,STCPmax,STC method reduces uncertainty to 60% of PR–PVUSA uncertainty for c-Si, on average.
• Two methods to calculate first year and initial performance loss rate are presented.

The interest in the assessment of performance loss rate (PLR) of Photovoltaic (PV) modules and arrays has been increasing as long as the global installed power expands and ages. Reliable performance metrics, statistical methods and filtering techniques exploiting continuous outdoor measurements are therefore needed, in order to foster solar bankability of PV systems. This work presents an improved estimation method to decrease the uncertainty associated to PLR   assessment by (a) using the array generated power metric corrected to Standard Test Conditions (STC), namely Pmax,STCPmax,STC, to minimize seasonal oscillations, (b) applying a filtering technique to eliminate outliers and (c) performing linear interpolation on Pmax,STCPmax,STC monthly averages series. Estimated PLR   and its uncertainty are assessed using three-years data from twenty-four grid-connected PV arrays representing nine different PV technologies and results are compared with two other widely-recognized performance metrics, namely: the Array Performance Ratio (PRaPRa) and the Array Photovoltaic for Utility Systems Applications (PVUSAaPVUSAa). Results show (a) that adding spectral correction to irradiance and temperature correction reduces the uncertainty of 25% on average and (b) that the uncertainty associated to Pmax,STCPmax,STC metric is reduced to more than 60% on average with respect to the other investigated metrics for crystalline silicon-based technologies, while it is comparable in the case of thin-film technologies. Finally, two procedures estimating the first year PLR and the the first five months PLR are presented and discussed.