Short-term power prediction method of photovoltaic power plant based on density peak hierarchical clustering

Abstract

The invention relates to a short-term power prediction method of a photovoltaic power plant based on density peak hierarchical clustering. The method comprises the following steps: (A) obtaining prediction day meteorological data, an SVM identification model for weather type identification, and power prediction models corresponding to different weather types; (B) according to the prediction day meteorological data, using the SVM identification model to carry out layer-by-layer identification, and determining the weather type of a prediction day; and (C) using the prediction day meteorological data as the input of the power prediction model of the corresponding weather type, and outputting a power predicted value corresponding to the prediction day; wherein the SVM identification model and the power prediction models are obtained in the following way: carrying out hierarchical clustering of historical day meteorological data to form K cluster sets each corresponding to a weather type, further establishing the SVM identification model for identifying weather types, and establishing the power prediction model based on the weather type at the same time. Compared with the prior art, the method has advantages such as high prediction precision.

Description

technical field [0001] The invention relates to a short-term power prediction method of a photovoltaic power station, in particular to a short-term power prediction method of a photovoltaic power station based on density peak hierarchical clustering. Background technique [0002] In recent years, with the development of the economy, the two major problems of energy shortage and environmental pollution have become increasingly prominent. These two problems are mainly caused by the consumption and extensive use of fossil energy. Therefore, the development of clean, efficient and pollution-free energy has become a solution to these two problems The most effective approach to big problems. Solar energy is a natural resource, which has time periodicity and proximity similarity, and is affected by factors such as temperature, humidity, wind speed and cloud cover, and has strong random mutation. The output power of the photovoltaic power generation system is mainly affected by cha...

AI Technical Summary

Problems solved by technology

However, due to the lack of certain types of labels in the historical collection data, the prediction accuracy is low

Images