摘要
针对Clara算法在电力工程大数据分析领域的应用问题,文中介绍了Clara算法的基本思路与算法步骤,进一步提出了基于Clara算法的电力工程造价评估方法。其采用Clara算法通过反复随机抽样方法,利用部分数据样本还原整体数据样本的特征,实现对大规模电力工程数据进行聚类分组,并将聚类结果作为多元回归分析(Multiple Regression Analysis,MRA)的输入数据,分析得到电力工程数据对其造价评估的影响模式。通过算例测算结果表明,相比于PAM聚类算法与K-means,所提Clara算法能够减小聚类分析的计算时间,同时提高电力工程造价评估的准确性。
This paper aims at the application of Clara algorithm in the field of power engineering big data analysis.This paper introduces the basic idea and algorithm steps of Clara algorithm,and further proposes the power engineering cost evaluation method based on Clara algorithm.The Clara algorithm repeatedly use random sampling methods to restore the characteristics of the entire data sample by using partial data samples.The large-scale data of power engineering are clustered and grouped,and the clustering results are used as input data for Multiple Regression Analysis(MRA)analyzing the influence mode of power engineering data on its cost evaluation.Compared with PAM clustering algorithm and K-means,Clara algorithm can reduce the calculation time of clustering analysis and improve the accuracy of power engineering cost evaluation.
作者
刘士李
万一荻
赵迎迎
何辉
刘大平
LIU Shili;WAN Yidi;ZHAO Yingying;HE Hui;LIU Daping(Economic Research Institute,State Grid Anhui Electric Power Co.,Ltd.,Hefei 230022,China;State Grid Anhui Electric Power Co.,Ltd.,Hefei 230022,China)
出处
《电子设计工程》
2021年第4期131-134,138,共5页
Electronic Design Engineering
基金
国网安徽省电力有限公司双创项目(C120900099)。
关键词
Clara算法
聚类
电力工程
造价评估
Clara algorithm
clustering
power engineering
cost evaluation
