本文已被:浏览 127次 下载 953次
投稿时间:2023-02-28 修订日期:2023-06-02
投稿时间:2023-02-28 修订日期:2023-06-02
中文摘要: 传统的数据聚类处理主要采用基于聚类中心的方式,但其存在一些限制,例如需要预先确定聚类中心的数量,并对数据的分布形态有一定的要求。针对这一问题,本论文选择基于密度聚类算法作为解决方案,重点研究了电力设备位置信息的聚类问题。在介绍密度聚类原理和常用算法的基础上,分析了电力设备位置信息的特点和处理方法,介绍了OPTICS、HDBSCAN和DBSCAN三种密度聚类算法的实现步骤,并与传统的K-mean聚类算法进行比较,通过实验设计和结果分析,验证了密度聚类方法的有效性和适用性。最后,通过应用案例分析,探讨了这些方法在电力系统分析中的应用实例和价值。研究结果表明,基于密度聚类算法的电力设备位置信息聚类方法可以有效地帮助电力系统实现数据的快速分析,具有重要的理论和应用价值。
Abstract:Traditional data clustering methods mainly use a centroid-based approach, but there are some limitations, such as the need to pre-determine the number of cluster centers and requirements on the data distribution. To address this issue, this paper chooses density clustering algorithms as the solution and focuses on studying the clustering problem of power equipment location information. Based on introducing the principles and common algorithms of density clustering, this paper analyzes the characteristics and processing methods of power equipment location information, introduces the implementation steps of three density clustering algorithms, OPTICS, HDBSCAN, and DBSCAN, and compares them with the traditional K-means algorithm. Through experimental design and result analysis, the effectiveness and applicability of density clustering methods are verified. Finally, through case analysis, the application examples and value of these methods in power system analysis are explored. The research results show that the clustering method of power equipment location information based on density clustering algorithms can effectively help the power system achieve rapid data analysis and has important theoretical and practical value.
keywords: density clustering location information power equipment data preprocessing cluster centers
文章编号: 中图分类号: 文献标志码:
基金项目:
作者 | 单位 | |
高翔 | 国网上海嘉定供电公司 | 252865437@qq.com |
贺静 | 国网上海嘉定供电公司 | 312056461@qq.com |
陆嘉铭 | 国网上海市电力公司 | |
张一彦 | 国网上海嘉定供电公司 | |
朱洪志 | 国网上海嘉定供电公司 | |
李丹戎 | 国网上海嘉定供电公司 |
引用文本: