Photovoltaic power interval forecasting method based on kernel density estimation algorithm

Photovoltaic(PV) power forecasting is of great significance to improve the access level of photovoltaic power generation. Deterministic forecasting methods often fail to meet the needs of grid risk analysis and decision-making, and a single model is also difficult to adapt to the changes of PV objects under different meteorological conditions. In order to solve the above problems, this paper describes the seasonal distribution characteristics of PV output fluctuations, and a method of PV power uncertainty forecasting based on seasonal classification is proposed. Compared with the parameter estimation method, the nonparametric kernel density estimation method does not need to make assumptions about the distribution of the prediction error, and can obtain more information about the actual distribution of the error, and has no fixed requirements for the various distribution characteristics of the error, it has strong adaptability.