基于多属性评判和模糊因果网络的110 kV变电站故障诊断

针对110 kV变电站故障诊断中非确定性因素较多和在线性要求高的特征,以实际110 kV变电站的保护信息为基础,根据保护的因果关系将设备分成设备故障、保护动作、开关跳闸或拒跳三类基础相关节点,结合设备家族缺陷、运行维护评价和状态检修评价三个属性指标,从而搭建多属性评估模型来确立模糊因果网络(Fuzzy Cause-Effect Networks,FCE-Nets)相关节点的因果系数。通过故障量相关函数完善遗漏的保护信息,提出基于多属性评判和FCE-Nets的变电站故障诊断方法。通过实际应用分析,对比现有方案的故障诊断结果,验证了此方法的有效性。

Fault diagnosis for 110 kV substation based on multi-attribute evaluation and fuzzy cause-effect networks

To deal with the uncertain factors and high real-time performance, the elements of 110 kV substation is divided into three kinds of basic correlated nodes by the cause-effect relationship between delay protection equipments based on the protection information, the three kinds of basic correlated nodes are component fault, protection action and breaker tripping or refusing to trip. And then multiple attribute evaluation model is established to determine the causality intensity of associated nodes in FCE-Nets. The correlation function of electric parameters in failure is used to improve the missing information and checking mechanism. Finally, a substation fault diagnosis scheme is proposed based on multiple attribute evaluation and FCE-Nets. With the actual substation application analysis, the proposed scheme is compared with the existing method of fault diagnosis and the results verify the effectiveness of the scheme.