作　　者： (陈鹏伟）; (陶顺）; (肖湘宁）; (李璐）; (王旭辉）;

机构地区： 新能源电力系统国家重点实验室（华北电力大学）,北京市昌平区102206

出　　处： 《中国电机工程学报》 2017年第16期4650-4661,共12页

摘　　要： 为定量评估配电网中不确定因素对各节点电压的不确定性影响,从线性定量描述的建模角度出发,提出一种基于弹性网估计的电压不确定水平分析方法。首先采用点估计法估计各节点电压期望值,并利用电压期望值构造分析样本;然后通过对分析样本的弹性网估计,确定各不确定因素中对各节点电压的主导影响因素及其电压影响系数。针对系统电压不确定水平分析需求,由不确定因素概率分布特性与电压影响系数构建节点–不确定因素相关矩阵,提出不确定因素影响力度量指标和节点电压受不确定影响风险指标,以及基于潜在语义索引思想和k-means聚类的节点分级分区划分方法。最后,基于IEEE 33节点系统的测试结果表明,所提弹性网分析方法及指标能够准确评估不确定因素对电压的不确定性影响。 In order to quantitatively assess the influence of uncertainties on node voltages in distribution network, a new analYSiS model with elastic net regression was proposed in this paper from the perspective of the linear quantitative description. The following procedure was to obtain the mode/ parameters defined as the voltage influence coefficients： firstly estimated the expected values of node voltage by pointestimate method, and conducted the analysis sample collection; then determined the dominant factors and calculated their voltage influence coefficients for the above sample by elastic net regression. To further analyze the voltage uncertainty level in the whole distribution network, this paper proposed two indicators about uncertain impact output level and uncertain impacted risk level, as well as the classification method for nodes based on the Latent Semantic Indexing and k-means clustering. The verification results based on IEEE 33-bus test system demonstrated that the analysis model with elastic net and indicators had an excellent performance and could be of high accuracy in assessing the influence of uncertainties on node voltage.

关 键 词： 配电网 电压水平 不确定影响 弹性网 奇异值分解