A power prediction model and its validation for a roof top photovoltaic power plant considering module degradation. (August 2021)

• Record Type: Journal Article
• Authors: Malik, Prashant; Chandel, Rahul; Chandel, Shyam Singh
• Is Part Of: Solar energy. Volume 224(2021)
• Page Start: 184
• Page End: 194
• Publication Date: 2021-08
• Subjects: Photovoltaics -- Roof top PV power plant -- Solar cell model -- PV degradation -- Solar power prediction -- Sun simulator; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2021.06.015
• Languages: English
• ISSNs: 0038-092X

Abstract

Highlights: A new solar model including power degradation of aging modules is developed for realistic power prediction of PV plants. Solar cell parameters are measured using a class A Sun simulator to determine degradation. The model is validated for a 1 kWp PV power plant under real outdoor conditions. Accuracy of the developed model is found to be better than existing models. Model can be used for output power prediction of PV plants for any location worldwide. Abstract: In this study, power generation prediction of a photovoltaic(PV) power plant is carried out using a new solar cell model which also include power degradation of aging modules. The solar cell parameters are measured using a class A Sun simulator to determine PV degradation. The predicted results are validated experimentally using automatic data monitoring system under real outdoor conditions. The model with three-parameters, is found to be more accurate with least mean absolute percentage error 3.7% under low solar irradiance range. The results for four and five parameter models are found to be more precise in medium (150–500 W/m 2 ) and high (>500 W/m 2 ) solar irradiance ranges with mean absolute percentage error of 5.4% and 4.21% respectively. The results show 2% decrement in maximum power of PV plant after 4 years of its installation due to module degradation. The results are also compared with previous models which show the present model has higher accuracy of about 26%. The impact of series and shunt resistances at different irradiance levels, is also discussed. The results reveal that present model after including degradation can predict the real output power of a PV system more accurately. Follow up research areas are also identified.