文章摘要
宋军英,何聪,李欣然,等.基于特征指标降维及熵权法的日负荷曲线聚类方法[J].电力系统自动化. DOI: 10.7500/AEPS20181115008.
SONG Junying,HE Cong,LI Xinran, et al.Daily Load Curve Clustering Method Based on Feature Index Dimension Reduction and Entropy Weight Method[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20181115008.
基于特征指标降维及熵权法的日负荷曲线聚类方法
Daily Load Curve Clustering Method Based on Feature Index Dimension Reduction and Entropy Weight Method
DOI:10.7500/AEPS20181115008
关键词: 特征指标降维  熵权法  加权FCM算法  负荷曲线聚类
KeyWords: feature index reduction  entropy weight method  weighted FCM algorithm  load curve clustering
上网日期:2019-05-15
基金项目:国家自然科学基金项目(51577056);湖南省电力公司资助项目(5216A0140090)
作者单位E-mail
宋军英 国网湖南省电力有限公司 276435879@qq.com 
何聪 湖南大学电气与信息工程学院 749127148@qq.com 
李欣然 湖南大学电气与信息工程学院 lixinran1013@qq.com 
刘志刚 国网湖南省电力有限公司 liuzg@hn.sgcc.com.cn 
汤杰 湖南大学电气与信息工程学院 2647532500@qq.com 
钟伟 国网湖南省电力有限公司 zhongw@hn.sgcc.com.cn 
摘要:
      日负荷曲线聚类是负荷建模背景下分析负荷特性的基础。针对现有聚类方法在聚类质量、聚类效率等方面的不足,综合运用模糊C均值及熵权法原理提出一种基于特征指标降维及熵权法的日负荷曲线聚类方法。首先提取日负荷率、日峰谷差率、日最大利用时间等7类降维特征指标替代各采样点负荷数据作为聚类输入;其次,引入熵权法自适应配置各特征指标的权重系数;最后,采用特征加权的模糊C均值聚类算法(FW-FCM)对用电日负荷曲线进行聚类。用本文所提方法对某地区日负荷曲线进行聚类分析,算例结果表明本文方法在运行效率、鲁棒性、聚类质量等方面有一定的优越性,聚类结果能真实有效的反映负荷的实际用电特性。
Abstract:
      Daily load curve clustering is the basis of analyzing load characteristics under the background of load modeling. Aiming at the shortcomings of existing clustering methods in clustering quality and efficiency.Based on the principle of fuzzy C-means and entropy weight method, a clustering method of daily load curve is proposed, which is based on dimensionality reduction of characteristic index and entropy weight method.Firstly, seven kinds of dimensionality reduction characteristic indexes such as daily load rate, daily peak-to-valley difference rate and daily maximum utilization time were extracted to replace the load data of each sampling point as a clustering input, Secondly, entropy weight method was introduced to adaptively allocate the weight coefficient of each characteristic index. Finally,a feature-weighted fuzzy C-means clustering algorithm (FW-FCM) was used to cluster the power load curves. The daily load curve of a region is clustered by using the proposed method in this paper. Results of calculation examples show that the method has certain advantages in operating efficiency, robustness, clustering quality, etc., and the clustering results can reflect the actual power consumption characteristics of the load truly and effectively.
查看全文   查看附录  查看附录  查看附录  查看附录   查看/发表评论  下载PDF阅读器