A photovoltaic power prediction method based on weather classification

Abstract

The invention relates to a photovoltaic power prediction method based on weather classification. The method comprises the following steps: S1, acquiring local historical data; s2, classifying historical data according to the total cloud amount in the data, classifying the data with the total cloud amount lower than or equal to 20% into a weather type 1, and further classifying the data with the total cloud amount greater than 20%; s3, for the corresponding part of data needing to be further classified, calculating to obtain a weather type index SCF by using the sunshine percentage, the direct incidence ratio and the corrected atmospheric clarity index in the data, and further dividing the SCF into a weather type 2, a weather type 3 and a weather type 4 according to the size range of the SCF; s4, performing input factor diagnosis analysis on the four weather types, and then selecting different input factors; s5, after principal component analysis dimensionality reduction is carried out on the input variables, using a statistical model for distinguishing weather types to carry out training modeling, and then photovoltaic output prediction is carried out. The invention has the advantages that the photovoltaic power prediction error is reduced, the precision is high, and the weather type identification effect is better.

Description

technical field [0001] The invention relates to the technical field of photovoltaic forecasting, in particular to a photovoltaic power forecasting method based on weather classification. Background technique [0002] Our country proposes to continue to promote the reform of energy structure, always adhere to the strategic direction of green and low carbon, focus on increasing the proportion of clean energy consumption, and gradually realize the replacement of the existing main energy, that is, fossil energy. The combustion of traditional fossil fuels is the main source of greenhouse gases, and the use of renewable energy to replace traditional fossil energy can effectively reduce greenhouse gas emissions. Under the tide of the era of green power development, my country's power industry is based on the characteristics of abundant energy resources such as wind energy and solar energy, and vigorously promotes the development of renewable energy, forming a situation of diversifi...

AI Technical Summary

Problems solved by technology

[0007] 1. Most of the existing weather classification methods choose a single index, and the meteorological conditions are more complicated and difficult to describe with a single index

[0008] 2. The selection of classification indicators in the existing weather classification methods lacks a unified standard

[0009] 3. The existing weather classification method is not clear enough to describe the physical meaning of each weather type, and the difference of photovoltaic output under different weather types is not obvious enough

Images