Modeling Of Photovoltaic Cells And Its Application In Energy Prediction

The use of photovoltaic cells has increased in the last few decades.Designers need a reliable PV model to design PV system, predict energy production and so on. We should become more concerned about the research of PV models.The behavior of photovoltaic (PV) cells can be modeled with an equivalent circuit including a series resistance and a diode in parallel with a shunt resistance.To make the design and optimization of solar energy systems convenient, it is very essential to choose the suitable model of engineering mathematics. A convenient and practicable mathematic model of silicon solar cell is offered in this paper. Simultaneously a formula for calculating maximum power point is also obtained.A few models that predict energy production have been developed, but they require a large amount of input data that are normally not available during the design phase,such as Sandia's PV arry model. The 5-Parameter model and Luft et.al's TRW model investigated in this research uses the data provided by the manufacturers and semi-empirical correlation equations to predict the energy production for specified cell parameters and operating conditions. Comparing Predicted I-V curves with the measured I-V values,there're errors.We make a correction on the 5-Parameter model and reduce the errors. We also use Luft et.al's TRW model to predict energy,compare with the measured and analyse the reasons that causing the errors.Researchers often use statistical methods such as artificial neural networks to establish models that predict energy production by the weather forecast.In the paper, Some meteorological elements that affect PV array'output are discussed. A novel photovoltaic power prediction module based on BP neutral network is designed, which is trained and simulated by historical power and weather data provided by experiment. The result of prediction shows that this model has a high accuracy.