基于模糊时间Petri网的电网故障诊断方法  

Fault Diagnosis Method of Power System Based on Fuzzy Time Petri Net

作　　者：徐彪[1] 尹项根[1] 汪旸 张哲[1] 郝洵[1] 陈钟钟 

机构地区：[1]强电磁工程与新技术国家重点实验室(华中科技大学),湖北省武汉市430074 [2]国家电网华中电力调控分中心,湖北省武汉市430077

出　　处：《电力系统自动化》2018年第2期70-76,共7页Automation of Electric Power Systems

基　　金：国家重点研发计划资助项目(2016YFB0900600)~~

摘　　要：故障元件判别作为电网故障应急处置的首要环节,是在线调控运行的重点工作之一。针对现有基于Petri网的电网故障诊断方法未充分利用时序信息或者时序推理过程复杂的问题,提出一种基于模糊时间Petri网的电网故障诊断方法。为Petri网中库所及变迁引入时间属性以表征电力系统告警信息的时序约束关系,定义了置信概率与时序约束的关联推理运算,并从模型结构出发建立了模糊时间Petri网的分层推理过程,能够同时推理得到元件故障的置信概率及其时间点约束。算例仿真表明,所提方法能够快速判定故障元件,诊断结果正确且合理,具有较强的容错性,且能对告警信息做出正确评价。As the first link of power system failure emergency treatment, fault section estimation plays an important role in power system online control operation. Since the existing Petri net based fault diagnosis method does not make full use of the timing information and temporal reasoning process is complicated and time-consuming, a fault diagnosis method based on fuzzy time Petri net is proposed. Firstly, a time interval is set for each place and transition of the Petri net model to describe the timing constraints of the alarm information. Then related reasoning operation of the confidence probability and the timing constraint is defined. And based on the model structure, the hierarchical reasoning process of the fuzzy time Petri net is established. The reasoning process can be used to infer the confidence probability of the component failure as well as its time point constraint. Finally, the simulation results show that the proposed method can quickly determine the faulty components with strong fault tolerance. The diagnosis results are correct and reasonable, which can make correct evaluation of the alarm information.

关 键 词：故障诊断 模糊时间PETRI网 时序推理 动作评价 

分 类 号：TM711[电气工程—电力系统及自动化]