东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2006, Vol. 27 ›› Issue (3): 276-279.DOI: -

• 论著 • 上一篇    下一篇

用于水电厂设备的故障诊断的贝叶斯网络模型

张晓丹;赵海;谢元芒;尹震宇;   

  1. 东北大学信息科学与工程学院;东北大学信息科学与工程学院;辽宁省交通厅运输管理局;东北大学信息科学与工程学院 辽宁沈阳110004;辽宁沈阳110004;辽宁沈阳110003;辽宁沈阳110004
  • 收稿日期:2013-06-23 修回日期:2013-06-23 出版日期:2006-03-15 发布日期:2013-06-23
  • 通讯作者: Zhang, X.-D.
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(69873007)

Bayesian network model for fault diagnosis of hydropower equipment

Zhang, Xiao-Dan (1); Zhao, Hai (1); Xie, Yuan-Mang (2); Yin, Zhen-Yu (1)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China; (2) Transportation Management Office, Communications Bureau of Liaoning Province, Shenyang 110003, China
  • Received:2013-06-23 Revised:2013-06-23 Online:2006-03-15 Published:2013-06-23
  • Contact: Zhang, X.-D.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

549

摘要： 通过分析水电厂设备故障诊断所面临的不确定性等问题以及当前常用诊断方法存在的局限性,研究基于贝叶斯网络的设备故障诊断方法,提出了适合于诊断问题的贝叶斯网络结构并阐述了基于贝叶斯网络故障诊断的团树推理算法.该方法综合考虑了多故障、故障症兆模糊以及电厂设备操作之间有依赖关系等情况.通过丰满水电仿真系统中水机调速器故障诊断的应用实例,证实了该方法在信息不确定条件下进行诊断决策的有效性和准确性.

关键词: 故障诊断, 贝叶斯网络, 不确定性推理, 团树传播, 水电仿真

Abstract: The fault diagnosis based on Bayesian network is studied by analyzing the uncertainty and limitation found in the existing fault diagnosis commonly used for hydropower equipment. A Bayesian network architecture available to such fault diagnosis is proposed with relevant uncertainty reasoning and clique tree propagation algorithm described. In the diagnosis method proposed many factors are taken into account, such as multiple-fault, fuzzy fault symptoms and the mutual dependency of different operation to be done with the equipment of a hydropower plant. With the fault diagnosis for a speed governor in Fengman Hydropower Station as example, a simulation test was carried out and verified the effectiveness and accuracy of the decision via the fault diagnosis under conditions of uncertain information.

中图分类号: