Seasonal performance comparison of three grid connected photovoltaic systems based on different technologies operating under the same conditions

• Published in: Solar energy 2017-03, Vol.144, p.798-807
• Main Authors: Wang, Huaxin, Muñoz-García, M.A., Moreda, G.P., Alonso-García, M.C.
• Format: Article
• Language: English
• Subjects: Air mass influence; Electricity generation; Grid connected; Light soaking effect; Performance evaluation; Photovoltaic cells; Solar energy; Thermal effect
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/j.solener.2017.02.006

[Display omitted] •Seasonal Pmax varied on a-Si +5.9% (thermal annealing effect) and −5.7% (light soaking effect).•The total three-year performance ratio of poly-Si was 1% higher than a-Si and CdTe.•Instantaneous temperature effect on a-Si maximum power was negligible.•Air mass dependence of a-Si and CdTe were −11.22%/AM and −4%/AM, respectively. Three grid connected photovoltaic systems based on different PV technologies (poly-Si, a-Si and CdTe), operating under the same conditions were analysed to determine their degradation and the variation in seasonal performance. In addition to the intrinsically different degradation of each technology through indoor tests at standard test condition (STC), a deeper analysis of the variables affecting the fluctuations of the performance was performed. The performance seasonal fluctuations resulting from the solar spectrum and module temperature was studied during a single year (December 2014 to December 2015). A notable solar air mass spectrum dependence (−11.22%/AM) of the generated power but scarcely detectable instantaneous temperature dependence was observed in the a-Si array. In contrast, the poly-Si array suffered the highest power decline with an elevating temperature (−0.56%/°C). The CdTe array exhibited a tempered interaction with the solar spectrum (−4%/AM) and temperature (−0.25%/°C). Finally, the significant maximum power difference between two months (April and September) due to the thermal annealing and light soaking effects in the a-Si array is confirmed, where the quantified variation of these two effects were 5.9% and −5.7%, respectively.