基于贝叶斯网电网多区域复杂故障诊断研究.pdfVIP

第 39 卷 第 7 期 电力系统保护与控制 Vol.39 No.7 2011 年 4 月 1 日 Power System Protection and Control Apr.1, 2011 基于贝叶斯网的电网多区域复杂故障诊断研究 宋功益，王晓茹，周 曙 （西南交通大学电气工程学院，四川 成都 610031） 摘要：针对大电网复杂拓扑结构的电网故障情况，先对其进行分割，提出一种基于粗糙集与贝叶斯网相融合的分层递归型诊 断网络模型。利用粗糙集理论的知识约简和处理不确定信息的能力，对电网故障诊断知识进行分层挖掘，实行属性优选，再 运用贝叶斯网络完成故障区域及故障元件的识别。该方法对复杂故障采用多区域并行诊断。算例结果表明，该方法正确、有 效，能提高系统丢失关键信息情况下的容错性，实时性也很高，具有很好的实用价值。 关键词：故障诊断；复杂故障；电网分割；粗糙集；贝叶斯；并行诊断 Research on complex faults diagnosis of multi-area power network based on Bayesian networks SONG Gong-yi，WANG Xiao-ru ，ZHOU Shu （School of Electric Engineering，Southwest Jiaotong University，Chengdu 610031，China ） Abstract ：According to the fault conditions of complicated topology and large-scale power networks，an efficient method is proposed to partition the large-scale power networks ．A hierarchical recursive fault diagnosis model is proposed based on rough set and Bayesian network. Using the ability of knowledge reduction and processing indeterminate information of rough set theory ，the hierarchical mining of substation’s fault diagnosis knowledge is carried out and optimal seeking of attributes is performed. Then Bayesian network is applied to identify fault areas and fault components ．For complicated fault, this paper adopts multi-area parallel diagnosis ．Results of calculation examples show that the proposed method is correct and effective ，and can improve the fault tolerance capability and speed of the fault diagnosis system while the kernel attribute is lost，so this method is available． Key words ：fault diagnosis ；complex faults ；network partition ；rough set ；Bayesian networks ；parallel diagnosis 中图分类号： TM77 文献标识码：A 文章编号： 1674-3415(2011)07-0020-06 学者把贝叶斯网络应用于不同领域的故障诊断。以 0 引言