考虑电站集群与发电时变特性的风光出力描述方法

Description of Wind and Solar Power Generation Considering Power Plant Clusters and Time-varying Power Characteristics

风光电站群“点多面广”及发电间歇性、波动性带来大规模并网消纳挑战,如何对风光电站合理分组并准确刻画风光集群出力特性是解决这一挑战的重要途径。依托云南电网实际工程,提出考虑电站集群与发电时变特性的风光出力描述方法。构建了电站群出力互补准则,采用凝聚层次聚类方法进行集群划分,刻画了互补性指标与集群数量二维关系,以确定合理的集群划分方式;引入时变特性,提出集群电站时序出力分段方差最小目标,利用分裂层次聚类确定最优分段及时段数量,建立了分段差异化集群出力概率密度分布函数。采用21座风、光电站实际工程数据进行验证,给出了单一类型电站集群和风光电站混合集群3个实例,结果表明90%置信区间的可靠性均超过92%,与单站调度方式相比,平均区间宽度分别减小29%、33%、43%,与全时段统一描述方式相比,前述区间宽度分别减小7%、34%、9%,呈现出较好的可靠性、集中度和实用性。

The widely-scattered wind and PV power plants and the intermittence and fluctuation of power generation have brought about great challenges to the large-scaled grid-connected consumption.It is an effective way to overcome these challenges how to group the wind and PV power plants reasonably and describe the generation characteristics accurately of the power plant clusters.Based on the actual project in Yunnan power grid,a method for describing wind and solar power generation considering the power plant clusters and the time-varying power characteristics is proposed.In this paper,we construct the output complementary criterion of the power plant clusters,adopt the agglomerative hierarchical clustering method for cluster division,and describe the two-dimensional relationship between the complementary index and the number of clusters to determine the reasonable cluster division.Then introducing the time-varying characteristics,we establish a different probability distribution density function of the generation of power plant clusters considering time-varying characteristics,setting the objective of minimizing the sectional variance of sequential output of the power plant clusters and determining the optimal number of the segments and the timely segments by splitting the hierarchical clustering.Taking 21 wind and PV power plants in this area as the engineering background,through the analysis of three examples with single-typed power plants and mixed-typed plants,we find that the descriptive reliability in 90%confidence interval is more than 92%.Compared with the single plant description,the average interval widths are reduced by 29%,33%and 43%,respectively.Compared with the whole time unified description,the widths of the aforementioned intervals is reduced by 7%,34%and 9%,respectively.The proposed description shows good reliability,concentration and practicability.