基于模糊C均值聚类的分布式电源集群并网逆变器输出功率置信区间估计方法

Estimation Method of DG Clusters Grid-connected Inverter's Output Power Confidence Interval Based on Fuzzy C-means Algorithm

为了提高多分布式电源仿真建模的精度,降低仿真模型与物理设备的运行误差,提出基于模糊C均值聚类的分布式电源并网逆变器输出功率置信区间估计方法。首先,根据DG的电压电流输入输出关系建立同类型DG的统一模型,采用粒子群算法进行参数辨识,得到DG输出功率偏差与物理系统同型设备运行参数的关系特性;其次,采用模糊C均值聚类方法,对DG集群进行分类,并将输出功率偏差划分成若干个子集;最后,采用概率分析方法得到输出功率偏差对应各子集的概率密度函数,并对偏差进行概率置信区间预测,根据预测结果通过物理系统同型设备输出特征修正仿真模型运行行为。采用实际运行算例验证,结果表明该方法快速准确修正了仿真模型与物理模型的运行误差。

In order to improve accuracy of multi-distributed generation simulation models and reduce the output error between the simulation model and physical device,an estimation method of output power confidence interval of distributed power supply grid-connected inverter based on fuzzy C-means algorithm is proposed.Firstly,a unified model of the same type of DG is established according to the input/output relationship of voltage and current of DG,and the parameter identification method based on particle swarm optimization algorithm is used to obtain the relationship between DG output power deviation and physical equipment’s operating parameters.Secondly,the fuzzy C-means algorithm method is used to classify DG clusters and divide the output power deviation into several subsets.Finally,the probabilistic analysis method is used to determine the probability density function of each subset of output power deviation,and then the probability interval prediction is performed on the deviation,and the simulation model’s operating behavior is corrected by physical equipment according to prediction result.Actual operation example shows that the method can quickly and accurately correct the operating error between the simulation model and actual operating equipment.