Method and system for power prediction of photovoltaic power station based on operating data of grid-connected inverters

Abstract

The present disclosure provides a method and system for power prediction of a photovoltaic power station based on operating data of grid-connected inverters, including: constructing a photovoltaic module model according to parameters of a photovoltaic module in a photovoltaic power station; constructing a power prediction model based on an artificial neural network algorithm; acquiring output data of a photovoltaic array when being shaded by static shadows of different thicknesses and different shading areas, constructing a training set to train the power prediction model, and obtaining a trained power prediction model; and acquiring, classifying, and normalizing output powers in real-time operating data of an inverter when the photovoltaic array is under a clear sky condition, and predicting a output power of the entire photovoltaic power station by using the trained power prediction model, the power prediction including a rolling prediction of the output power of the photovoltaic power station under a clear sky condition and a minute-level power prediction of the photovoltaic power station when being shaded by a dynamic cloud cluster. The present disclosure reduces a device cost and overcomes the defect that cloud clusters of different thicknesses affect the precision of power prediction of the photovoltaic array.