多因素分级Petri网配电故障诊断方法  

Distribution Network Fault Diagnosis Method Based on Multi-Factor Hierarchical Petri Net

作　　者：贾学涵 杨东升[1] 郑薇 毕影娇 

机构地区：[1]东北大学信息科学与工程学院,辽宁省沈阳市110819 [2]国网营口供电公司,辽宁省营口市115000

出　　处：《电网技术》2017年第3期1015-1021,共7页Power System Technology

基　　金：国家自然科学基金项目(61273029)~~

摘　　要：针对采用概率方法进行故障诊断易造成调度操作风险大的问题,提出一种配电网故障诊断的多因素分级Petri网模型,主要解决断路器或保护拒动、误动等故障信息不确定情况下的故障问题。该方法充分考虑了保护启动信息、保护出口信息、重合闸动作信息、断路器动作信息、事件顺序记录(sequence of event,SOE)时序信息以及连续电气量,实现故障信息不确定情况下的电网故障诊断。当电网故障存在信息不确定情况时,首先将连续电气量和SOE时序信息作为该模型的一级诊断区的输入,然后通过将连续电气量的故障时刻与SOE时序的故障时刻进行比较,根据比较结果判断并激发托肯能量变化的推理过程,给出存在信息不确定的情况下电网故障诊断结果。最后,针对某局部配电网模型进行了仿真验证,结果表明了该方法的有效性。Using probabilistic methods in fault diagnosis can bring a significant risk for dispatching operation. This paper presents a multi-factor hierarchical Petri net model for distribution network fault diagnosis to solve problem of mal-operation or failure to trip protection relay and circuit breaker when fault diagnosis information is uncertain. This method took fully into account information of protection starting, protection exiting, reclosing and circuit breaker action, sequence of event（SOE） timing and continuous electrical quantity and realized distribution network fault diagnosis under uncertain fault information. When fault information of distribution network was uncertain, this method would took continuous electrical quantities and SOE timing information as input of primary diagnosis zone of the model. Then, reasoning process for determining and stimulating Token energy＇s change was presented according comparison results of fault time between continuous electric quantity and SOE timing information, and fault diagnosis result under uncertain information was obtained. Finally, simulation was performed for a local distribution network model with results showing effectiveness of the proposed method.

关 键 词：保护启动信息 连续电气信息 多因素分级Petri网 托肯能量 SOE 

分 类 号：TM744[电气工程—电力系统及自动化]