摘要
随着光伏电源大规模入网,光伏出力的不确定性给电网规划运行带来了挑战。针对流域光伏集群出力不确定性的描述问题,引入参数离心系数与均匀缩放系数改进(density-based spatial clustering of application with noise,DBSCAN)算法,对出力数据去噪重构,使用Kmeans++算法对去噪后数据进行场景提取,基于此提出了一种耦合DBSCAN-Kmeans++算法的流域光伏集群出力的场景提取方法。此方法应用于某省A、B、C 3个流域的实例研究表明,三江枯水期场景提取的F值较原始方法分别提升了57.68%、182.74%、57.41%。研究生成了三江丰枯期光伏出力场景集,提取结果与光伏年内出力特性相符,对比三江出力场景,表明B流域的光伏效能更优。所提出的方法可用于有噪数据环境下多站尺度的光伏出力不确定性描述。
As photovoltaic power sources are increasingly integrated into the grid,the uncertainty of their output presents significant challenges to the planning and operation of the power grid.To describe the uncertainties in the output of photovoltaic clusters in river basins,this paper introduces the concepts of parameter eccentricity coefficient and uniform scaling coefficient to enhance the DBSCAN algorithm.The output data is denoised and reconstructed,and the Kmeans++algorithm is utilized to extract scenes from the denoised data.On this basis,a scene extraction method for river basin photovoltaic cluster output,which combines the DBSCAN-Kmeans++algorithm,is proposed.A case study applying this method to three rivers basins—the A River,the B River,and the C River in a province—reveals that the F value extracted from the dry season scenes of these rivers has increased by 57.68%,182.74%,and 57.41%respectively,compared to the original method.The study has generated a set of photovoltaic output scenarios for the flood and dry seasons of the three rivers,and the extracted results align with the annual output characteristics of photovoltaics.By comparing the output scenarios of the three rivers,it is determined that the photovoltaic efficiency in the B River is superior.The method proposed in this article can be employed to describe the uncertainty of photovoltaic output at multiple station scales within noisy data environments.
作者
张阳博
哈西
马光文
陈仕军
黄炜斌
朱燕梅
ZHANG Yangbo;HA Xi;MA Guangwen;CHEN Shijun;HUANG Weibin;ZHU Yanmei(School of Water Resource and Hydropower,Sichuan University,Chengdu 610065,Sichuan,China;Yalong River Basin Hydropower Development Company,Chengdu 610056,Sichuan,China)
出处
《电网与清洁能源》
北大核心
2025年第7期132-140,共9页
Power System and Clean Energy
基金
四川省博士后科研项目特别资助(TB2023058)。
