Research On Generating Power Forecasting Of Grid-connected PV System

As the global energy shortage and security issues become increasingly prominent, the development of new energy and renewable resources to achieve sustainable economic development has become the consensus around the whole world. Among distributed generation systems, the technology of grid-connected solar photovoltaic power generation turns out to be an important research field. It has great application value in both civil and industrial electricity. Recent research has yielded some achievements, while in some key technologies, there are still problems to be solved.In the planning of grid development with grid-connected photovoltaic power generation systems, the implications of uncertainty on PV generation output when being integrated to power grid should be fully considered. This paper has a deep study on the prediction of grid-connected photovoltaic power generation, and establishes an analysis model based on multiple linear regressions, in order to predict the PV power generarion under different weather conditions. This will help to understand the operating characteristics of grid-connected solar photovoltaic generation systems, be beneficial to the grid scheduling and dispatching management, and will effectively reduce the random effects of PV generation, so as to improve the security and stability of power systems.On account of PV system generating uncertainty and periodicity, the paper analyzes the factors influencing the output of the PV system, raises and resolves a methods for PV power generation forecasting based on multiple linear regression model. On the basis of the pre-trial operation data of Alstom grid-connected PV system projects, parameter estimations and statistics tests according to the prediction model on three different weather conditions are analyzed. By using the regression equations, the generated power of PV systems are predicted, compared with the actual measurement data by SCADA. The results verify that the model was efficient and reliable with high accuracy.