Evaluation and improvement of the single diode model's parameters used for solar PV modelling

[摘要] ENGLISH ABSTRACT: Solar photovoltaics are one of the leading renewable energy generators that are being installed in South Africa. The understanding of how this technology works and how it responds to radiation, is important. This thesis evaluates the different parameters that influence the performance of a solar cell. The parameter extraction method is of interest as it tells the story of what is happening internally in a solar cell when subjected to environmental conditions. The response of the cell to certain light is part of this study to evaluate the influence of different light and how it relates to the testing condition on the data sheet of the solar module.The solar cell parameters of the single diode equivalent circuit model are studied. The data originate from a testing facility in the Northern Cape. The data were examined to extract the five parameters from the current voltage curve, measured in 10 minute intervals on the site. The parameter extraction method of De Soto is presented and implemented to compare the induced errors made with this method. The error was anoverestimation of 13.71% in the maximum power point. The evaluation of the error resulted in an observation of a linear error with regards to temperature and irradiance.The output of the method was adjusted with a function to result in an improvement of a percentage error of -2,99%. The adjustment function was optimised and the errorin extracting the maximum power point was reduced to only -0.062%. The individualparameters are examined and adjusted, to present five new equations for the parameter extraction that are developed from De Soto's method. The new equations have shown an increase in parameter accuracy. When the new parameter extraction equations are usedtogether, the maximum power point error is -1.73%.An automated spectrum and solar module performance monitoring system was developed.The system measures the current voltage curve from the solar module with the solarspectrum at an interval of 10 minutes. The data are stored on a local and web-baseddatabase for analysis. The data were firstly analysed in the different bands of wavelength that each colour in the spectrum occupies. This has shown that red light is the dominantcolour in the spectrum that was measured. The relationship between the average photonenergy and the current and voltage of the module are examined and have shown that thevoltage is high regardless of the amount of energy, as long as there is direct sunlight onthe module. The current increases as the average photon energy quantification parameterincreases.From this study it is evident that the spectrum also plays a role in the modelling ofphotovoltaics, especially when the conditions are not standard test conditions. The solarspectrum can change considerably and it is important to take it into consideration whenaccurate modelling of photovoltaics is conducted.

[发布机构] Stellenbosch University