电工技术学报  2019, Vol. 34 Issue (14): 3012-3024    DOI: 10.19595/j.cnki.1000-6753.tces.180742
电力系统分析与应用 |
利用多因素降维的配电网区段定位完全解析模型
王秋杰1, 金涛1, 申涛2, 朱绍杰3, 刘随阳3
1. 福州大学新能源发电与电能变换福建省高校重点实验室 福州 350116 ;
2. 国网湖北电力有限公司随州供电公司 随州 441300 ;
3. 国网湖北电力有限公司荆门供电公司 荆门 448000
A Complete Analytic Model of Section Location in Distribution Network Based on Multi-Factor Dimensionality Deduction
Wang Qiujie1, Jin Tao1, Shen Tao2, Zhu Shaojie3, Liu Suiyang3
1. Fujian Key Laboratory of New Energy Generation and Power ConversionFuzhou University Fuzhou 350116 China;
2. State Grid Hubei Electric Power Co. Ltd Suizhou Power Supply CompanySuizhou 441300 China;
3. State Grid Hubei Electric Power Co. Ltd Jingmen Power Supply Company Jingmen 448000 China
全文: PDF (13133 KB)  
输出: BibTeX | EndNote (RIS)      
摘要 现有配电网区段定位模型由于没有将故障信息的漏报和误报计入故障变量,导致区段定位的容错能力在多节点信息畸变时存在不足。对此,构建一种配电网区段定位的完全解析模型来提高区段定位的容错能力。为了降低完全解析模型的变量维度,首先利用故障矛盾假说,将变量维度减小到3倍的节点数量;其次,利用模型分层,将单次区段定位的变量维度进一步减小;最后,将基于模型诊断的结果作为第一层定位的已知量,变量维度再次减小。算例表明:与未考虑漏报和误报的模型相比,该完全解析模型容错能力得到极大的提高,并且能同步获得漏报、误报警告信息;与没有利用多因素降维的完全解析模型相比,所提模型的定位速度和准确率相对较高,更加符合定位实时性要求。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
王秋杰
金涛
申涛
朱绍杰
刘随阳
关键词 配电网区段定位完全解析模型多因素降维    
Abstract:The existing section location model of distribution network does not count the information missing and false alarms into fault variables, resulting in the inadequacy of fault tolerance of section location when multi-node information is distorted. In order to improve the fault tolerance of section location, a complete analytical model of section location in distribution network is constructed. In order to reduce the variable dimension of the complete analytic model, the fault contradiction hypothesis is used to reduce the variable dimension to three times the number of nodes. Secondly, the variable dimension of single fault location is further reduced by a hierarchical location model. Finally, when the diagnosis results, obtained by the method of model based diagnosis, act as the known variables of the first layer model, the variable dimension is reduced again. The case shows that compared with the models that do not take account of missing and false alarms, the fault-tolerant capability of the complete analytic model is greatly improved, and the warning message of missing and false alarm can be obtained synchronously. Compared with the complete analytical models that do not use multi-factor to reduce the dimension, the speed and accuracy of the proposed model are relatively high, which is more in line with the real-time requirements of section location.
Key wordsDistribution network    section location    complete analytic model    multi-factor dimen- sionality deduction   
收稿日期: 2018-05-01      出版日期: 2019-07-29
PACS: TM72  
  TM77  
基金资助:欧盟FP7国际科技合作基金(909880)和国家自然科学基金(61304260)资助项目
通讯作者: 金涛 男,1976年生,教授,博士生导师,研究方向为电力系统稳定性分析、在线测量与信号处理、新能源技术。E-mail:jintly@fzu.edu.cn   
作者简介: 王秋杰 男,1988年生,博士研究生,研究方向为电力人工智能、配电网故障诊断。E-mail:84106438@qq.com
引用本文:   
王秋杰, 金涛, 申涛, 朱绍杰, 刘随阳. 利用多因素降维的配电网区段定位完全解析模型[J]. 电工技术学报, 2019, 34(14): 3012-3024. Wang Qiujie, Jin Tao, Shen Tao, Zhu Shaojie, Liu Suiyang. A Complete Analytic Model of Section Location in Distribution Network Based on Multi-Factor Dimensionality Deduction. Transactions of China Electrotechnical Society, 2019, 34(14): 3012-3024.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.180742          https://dgjsxb.ces-transaction.com/CN/Y2019/V34/I14/3012