基于告警时序和多源信息的输电设备故障诊断
DOI:
作者:
作者单位:

1.东北石油大学;2.新疆油田公司工程技术研究院

作者简介:

通讯作者:

中图分类号:

基金项目:

黑龙江省科学基金项目


fault diagnose of transmission device based on time series of alarm and multi-resource information
Author:
Affiliation:

1.Northeast Petroleum University;2.Engineering technology Research institute of XinJiang oilfield company

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    摘要:针对不计告警时序信息导致电网故障设备定位精度低和出现多解的问题和单个信息源的故障设备定位方法往往有不精确性和容错性不高的缺点,采用告警时序矩阵和多源信息结合的方法求解故障设备和故障时刻。首先基于告警时序信息,定义了一系列告警时序矩阵。构建了故障设备定位的时序目标函数。利用基于记忆的免疫遗传算法作为求解时序目标函数的方法。其次,采用数据采集与监控系统(SCADA)系统、故障录波信息系统、继电保护系统和广域测量系统(WAMS)系统的多源信息进行线路故障设备定位的方法。求出了时序矩阵法、保护值法、保护路径法、录波器法和WAMS法的故障概率分配函数。最后,采用基于模糊层次分析法(FAHP)改进D-S证据理论进行多源信息的融合。采用基于模糊层次分析法确定重要性权值的折扣证据法,解决了重要性权值不能进行理论性定量确定的问题。采用可能性C均值聚类算法来进行融合分类的方法,求出了故障设备类和非故障设备类的两个大类包含的故障和非故障设备。进行了验证和应用。结果表明,基于告警时序和多源信息的输电网定位方法可以完全识别系统的故障设备,识别结果可靠。

    Abstract:

    The fault location in power grid that ignores alarm timing information always has the problems of low positioning accuracy and appears several solutions。The fault location method by Single information source is inaccuracy,and the fault tolerance is low. Therefore, a new fault location method is proposed. With the alarm timing information,the alarm timing matrix and timing objective function for fault location are formed firstly. The timing objective function is formed by immune genetic algorithm. Then, with the SACDA system, fault recorder information system, relaying protection system and WAMS system, the distribute functions to the probability of the fault state are formed by time series matrix method, protection value method, protection path method, recorder method and WAMS method. Finally, the multi-source information is formed by improved D-S theory based on fuzzy analytical hierarchy process (FAHP). The discount evidence method based on fuzzy analytical hierarchy process is used to solve the problem of importance weights that can’t be determined quantitatively by theory. Possibility C mean clustering algorithm is used to locate the fault devices and non-fault devices. The results from the simulation system indicate that this method can locate the faults efficiently, accuracy.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-03-18
  • 最后修改日期:2021-06-15
  • 录用日期:2021-07-26
  • 在线发布日期:
  • 出版日期: