基于曲线形态特征的地区规模化风电出力场景划分  

Regional Scaled Wind Power Output Scene Segmentation Based on Curve Morphological Features

作　　者：林俐 肖舒 费宏运 潘险险 LIN Li;XIAO Shu;FEI Hongyun;PAN Xianxian(State Key Laboratory of New Energy Power Systems,School of Electrical and Electronic Engineering,North China Electric Power University,Beijing 102206,China;Dongguan Power Supply Bureau,Guangdong Power Grid Corporation,Dongguan 523000,Guangdong,China;Power Grid Planning Center of Guangdong Power Grid Corporation,Guangzhou 510080,Guangdong,China)

机构地区：[1]华北电力大学电气与电子工程学院新能源电力系统国家重点实验室,北京102206 [2]广东电网有限责任公司东莞供电局,广东东莞523000 [3]广东电网有限责任公司电网规划研究中心,广东广州510080

出　　处：《电网与清洁能源》2020年第3期74-81,88,共9页Power System and Clean Energy

基　　金：国家高技术研究发展计划(863计划)项目(2011AA05A104)。

摘　　要：目前多采用基于欧式距离的聚类方法对风电出力场景进行聚类划分,其结果反映时间序列曲线的幅度大小差异,而未反映曲线的形态特征及变化趋势的不同。据此,文中提出一种基于曲线形态特征的地区规模化风电出力场景划分方法。针对地区规模化风电出力时间序列曲线,定义考虑序列互相关性的“形态距离(shape-based distance,SBD)”,将其作为聚类算法的相似性度量函数,进一步给出基于形态距离的时间序列样本聚类划分过程。然后以某含规模化风电的实际地区为研究对象,提取该地区春季风电出力典型场景,并与传统K-means聚类算法的结果进行对比分析,验证所提方法的有效性。最后运用文中方法提取了该地区各个季节不同风况下的风电出力典型场景。At present,the clustering method based on Euclidean distance is frequently used to classify the wind power output scenarios,and its results reflect the magnitude of the time series curve,but not the morphological characteristics and trends of the sequence.To this end,this paper proposes a regional scaled wind power output scene segmentation method based on curve morphological features.For the regional scaled wind power output time series curve,the"Shape-Based distance(SBD)"is defined considering the sequence cross-correlation as a similarity measure function of clustering algorithm.Furthermore,the time series sample clustering process based on morphological distance is given.In addition,taking an actual area containing large-scale wind power as the research object,the typical scene of wind power output is extracted and compared with the results of the traditional K-means clustering algorithm to verify the effectiveness of the proposed method.Finally,the typical wind power output scenarios under different wind conditions in different seasons in this region are extracted with the method proposed in this paper.

关 键 词：规模化风电 聚类划分 典型场景 形态距离 

分 类 号：TM614[电气工程—电力系统及自动化]