容纳时序约束的改进模糊Petri网故障诊断模型

发布时间：2018-05-03 20:28

本文选题：电力系统 + 故障诊断　；参考：《电力系统自动化》2014年05期

【摘要】：现有基于Petri网理论的电力系统故障诊断模型与方法完全没有利用或没有充分利用警报信息的时序属性,且容错性不强,在处理复杂故障时可能无法得到明确的诊断结果。在此背景下,提出一种融合时序约束网络的模糊Petri网故障诊断模型。通过将Petri网的库所赋予时间属性,以充分计及元件故障、保护动作和断路器跳闸间的延时约束,并采用模糊加权算法进行推理运算,提高了模型的容错性。最后,以IEEE 10机39节点电力系统为例验证了所发展的模型的正确性和求解算法的有效性。与现有两种方法的比较表明,所发展的诊断方法能够自动跟踪网络拓扑变化,在增加少量计算量的情况下,较大提高了容错能力,满足大规模电力系统在线故障诊断需要。
[Abstract]:The existing power system fault diagnosis models and methods based on Petri net theory do not utilize or make full use of the time series attributes of alarm information, and the fault tolerance is not strong, so it may not be able to obtain definite diagnosis results when dealing with complex faults. Under this background, a fault diagnosis model of fuzzy Petri nets is proposed. By assigning the time attribute of the Petri net library to fully consider the delay constraints between component faults, protection actions and circuit breakers tripping, and using fuzzy weighting algorithm for reasoning operation, the fault tolerance of the model is improved. Finally, the correctness of the developed model and the validity of the algorithm are verified by taking the IEEE 10 machine 39 bus power system as an example. The comparison with the existing two methods shows that the developed diagnosis method can automatically track the network topology changes, and the fault tolerance ability is greatly improved with a small amount of computation, which can meet the needs of large-scale power system on-line fault diagnosis.
【作者单位】：浙江大学电气工程学院;浙江电力调度控制中心;
【基金】：国家科技支撑计划资助项目(2011BAA07B02) 浙江省电力公司重点科技项目~~
【分类号】：TM73

【参考文献】