Prediction method and simulation study of port photovoltaic power generation output power

Photovoltaic power generation has the characteristics of large randomness, poor stability, and great influence by meteorological conditions. A large number of grid-connected photovoltaic power system will bring random impact to the power grid, which may affect the stablility of the power grid. In this paper, at first we discuss the direct prediction method and indirect prediction method commonly used in current power prediction. Then the prediction methods including Vector Autoregressive (VAR) model and Light Gradient Bosting Machine(LGBM) model are applied to photovoltaic power prediction, the prediction simulation results of the two models are evaluated, which provides a reference for the research of photovoltaic power prediction method.