Combination weight similarity day selection-based photovoltaic power prediction method

Abstract

The invention discloses a combination weight similarity day selection-based photovoltaic power prediction method and belongs to the photovoltaic power generation technical field. According to the method, based on the basic value point errors of similarity variables, similarity errors are obtained through using a combination weight method, so that similarity days can be determined; the output power of the similarity days is weighted according to different weight coefficients, so that predictive power can be obtained; based on the principle of minimum discrimination information, a subjective weight and an objective entropy weight are combined effectively, so that a combination weight coefficient for similarity day selection can be obtained; and the power weight coefficients are generated according to similarity indexes. As indicated by measured data example simulation of a certain photovoltaic power station, the method can select similarity days with high similarity, improve the prediction precision of photovoltaic output power and benefit grid-connected operation of a photovoltaic power generation system and safe and economic dispatch of a power grid.

Description

technical field [0001] The invention relates to a photovoltaic power prediction method based on combination weight similar day selection, which belongs to the technical field of photovoltaic power generation. Background technique [0002] Photovoltaic grid-connected power generation has become the mainstream trend of photovoltaic power generation. However, photovoltaic power generation has shortcomings such as volatility and intermittency, and large-scale grid connection of photovoltaic power generation will affect the economical, safe and stable operation of the power system. Accurate prediction of output power of photovoltaic power generation can provide a useful reference for grid power dispatching. [0003] At present, photovoltaic power forecasting methods can be divided into two categories: physical methods and statistical methods. The physical method takes meteorological forecast data as input and uses physical equations for forecasting; the statistical method condu...

AI Technical Summary

Problems solved by technology

However, photovoltaic power generation has shortcomings such as volatility and intermittency, and large-scale grid connection of photovoltaic power generation will affect the economical, safe and stable operation of the power system.

Images