| 引用本文: | 吴素我,张焰,苏运.基于配用电数据关联的中压配电网断线故障诊断方法[J].电力自动化设备,2017,37(7): |
| WU Suwo,ZHANG Yan,SU Yun.Open-line fault diagnosis based on data association of MV distribution network[J].Electric Power Automation Equipment,2017,37(7): |
|
| 摘要: |
| 基于配用电信息系统数据和关联规则算法，提出一种诊断中压配电网分支线断线不接地故障的方法。通过分析相互关联的配用电信息系统数据，提出基于数据特征选择的关联规则挖掘方法，并通过卡方分裂算法将连续型特征量转换为布尔型特征量，同时采用MSApriori算法解决故障信息中的稀有项问题，然后在此基础上应用kulc准则消除冗余规则以形成约简的代表规则家族。以华东某地区配用电信息系统中的历史数据为依据进行实际算例分析，结果说明所提出的方法能够大量减少无效挖掘，显著提高效率和准确度，适用于中压配电网断线故障的在线诊断。 |
| 关键词: 配电网 数据关联 特征选择 布尔离散化 断线故障诊断 |
| DOI：10.16081/j.issn.1006-6047.2017.07.015 |
| 分类号:TM72 |
| 基金项目:国家电网公司科技项目(52094014001Q) |
|
| Open-line fault diagnosis based on data association of MV distribution network |
|
WU Suwo1, ZHANG Yan1, SU Yun2
|
|
1.Department of Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;2.State Grid Shanghai Municipal Electric Power Company, Shanghai 200437, China
|
| Abstract: |
| A method based on the data of its power distribution information system and the association rules is proposed for diagnosing the non-grounding open-line faults of MV distribution network. The correlated data of the electric power distribution information system are analyzed and a method of association rule mining based on the data feature selection is proposed. Chi square splitting algorithm is applied to convert the continuous features into Boolean features, the MSApriori algorithm is applied to solve the rare item problem of fault information, and kulc principle is adopted to eliminate the redundant rules and form a reduced rule family. A practical analysis is carried out based on the historical data of a regional power distribution information system in East China, which shows that, since the invalid data mining is greatly reduced and the efficiency and accuracy are significantly enhanced, the proposed method is suitable for the on-line diagnosis of open-line faults of MV distribution network. |
| Key words: distribution network data association feature selection Boolean discretization open-line fault diagnosis |
