基于马氏距离相似度量的光伏功率超短期预测方法的研究

提高光伏功率超短期预测精度可有效减小光伏发电并网对电力系统稳定性的影响。文章提出了一种基于马氏距离相似度量的光伏功率超短期预测方法。首先,文章采用Elkan K-means聚类分析方法对天气类型进行划分,并通过计算各气象因素与光伏电站输出功率间的灰色关联度,选出不同天气类型下影响光伏功率的主要气象因素;然后,根据样本日和预测日间主要气象因素的马氏距离选择若干个相似日,并将相似日的光伏功率作为预测模型的训练集,对预测日的光伏功率进行超短期预测。模拟结果表明:基于马氏距离相似度量得到的相似日光伏功率和预测日的相似度较高;将基于马氏距离相似度量得到的相似日光伏功率作为预测模型的训练集,可以提高光伏功率超短期预测精度,为光伏功率预测领域提供了有效的方法。

Ultra-short-term prediction of PV power based on similar days of Mahalanobis distance

Due to the change of day and night and the impact of climate uncertainty,PV power has the characteristics of randomness,intermittent,volatility and periodicity.Large-scale PV grid connection will affect the safe and stable operation of the grid.Therefore,this paper proposes an ultra-short-term prediction method of PV power based on Mahalanobis distance similarity metric.First,the Elkan K-means cluster analysis method is used to classify the weather types;The main factors which affecting the PV power under the corresponding weather type are selected by calculating the gray correlation degree between various meteorological factors and the output power of the PV power station.Then,several similar days are selected according to the Mahalanobis distance between the sample day and the main meteorological factors during the forecast day.The sample data of the similar days and the numerical weather forecast data of the forecast day are input into the forecast model to conduct ultra-short-term forecasting of PV power.The research results show that the similarity date obtained based on the Mahalanobis distance similarity metric has a high degree of similarity with the PV power of the forecast date.Inputting it as a training set into the prediction model can improve the accuracy of the ultra-short-term PV power prediction,providing a great opportunity for the PV power prediction field.effective method.