基于逐步聚类分析的短期光伏发电预测方法

光伏发电功率与气象因素密切相关,可靠的功率预测对光伏入网和电网安全运行具有重要意义。为提高光伏短期发电功率预测的准确率,基于某40 MW光伏电站历史功率和气象数据,在不同季节和天气类型下利用逐步聚类分析方法(SCA)搭建光伏短期预测模型,实现分季节和天气类型的光伏功率预测。模型对比结果表明：逐步聚类分析方法具有较高的预测精度,在四季、单一天气类型和复合天气类型3方面预测精度分别提高了11.13%,9.51%和8.26%。

Short-term Photovoltaic Power Generation Prediction Method based on Stepwise Clustering Analysis

Photovoltaic power is closely related to meteorological factors, and reliable power prediction is of great significance for photovoltaic grid connection and safe operation of power grid. In order to improve the accuracy of short term photovoltaic power forecast, based on the historical power and meteorological data of a 40 MW PV power station, under different seasons and the weather types using the method of stepwise clustering analysis (SCA), short term photovoltaic forecasting model was built to implement photovoltaic power prediction classified by season and weather types. The results show that the stepwise clustering analysis method has high prediction accuracy, and the prediction accuracies of four seasons, single weather type and composite weather type are improved by 11.13%, 9.51% and 8.26%, respectively.