基于多权重混合分布模型的光伏出力波动特性研究

STUDY ON FLUCTUATION CHARACTERISTICS OF PV POWER OUTPUT BASED ON MULTI-WEIGHT MIXED DISTRIBUTION MODEL

建立光伏发电理论出力模型,通过求取其实际出力和衰减理论出力差值提取光伏出力的随机性分量,采用不同单一概率密度函数对随机性分量进行拟合,并分析拟合结果存在较大偏差的原因。为降低拟合偏差,采用基于二分K均值聚类的权重值划分方法,提出三权重混合t Location-scale概率密度函数模型以描述光伏出力波动量的数字特征。最后通过Matlab仿真结果与指标值验证所提出模型拟合效果的优异性。

Accurate establishment of the photovoltaic(PV)output model and description of its fluctuation characteristics play an essential role in PV grid-connected operation planning.This paper establishes the theoretical output model of PV power generation.It extracts the stochastic components of PV output by calculating the actual output and the attenuation theory output difference.Furthermore,it uses different single probability density functions to fit the stochastic components.Finally,the paper and analyses the reasons for the large deviation of the fitting results.In order to reduce the fitting deviation,a three-weight mixed t location-scale probability density function model is proposed to describe the digital characteristics of PV output fluctuation by using the weight partition method based on bisecting K-means clustering.Finally,the simulation results of Matlab and the index values verify the excellent fitting effect of the proposed model.