基于初始化非负矩阵分解的光伏发电预测

光伏发电作为一种可再生能源发电技术,使用规模日益扩大,光伏发电的精准预测已成为数据挖掘的重要研究领域,但是光伏发电本身的随机性和不确定性使得现有的预测功率难以达到理想的高度,对电网运行的稳定性造成了不利的影响.为了探索预测日的发电功率、历史日的发电功率以及气象数据等相关因素之间的关系,解决光伏发电难以准确预测的问题,提...

Prediction of Photovoltaic Power Generation Based on Initialized Non-negative Matrix Factorization

As a renewable energy power generation technology,photovoltaic power generation is increasingly used.The accurate prediction of photovoltaic power generation has become an important research area for data mining.However,the randomness and uncertainty of photovoltaic power generation makes it difficult for the current predicted power to reach the desired height,which adversely affects the stability of grid operation.In order to explore the relationship between the forecasted generation power and historical generation power,meteorological data and other related factors and to solve the problem that photovoltaic power generation is difficult to accurately predict,a photovoltaic power generation prediction method based on non-negative matrix factorization(NMF)is proposed.This algorithm not only establishes the relationship between the factors affecting photovoltaic power generation,but also provides a better initial value for the construction of the prediction model based on the fuzzy C-means(FCM).Applying this model to a real data set,the two weather conditions were predicted from 7 am to 6:15 pm for 46 sample points.The comparison experiments with the basic NMF models show that the model can effectively improve the accuracy of photovoltaic power generation prediction.