一种基于多参量融合故障树的变压器故障诊断方法

A multi parameter fusion methods for transformer fault diagnosis based on fault tree

本文提出了一种基于多参量融合故障树的变压器故障诊断方法。首先收集整理变压器故障诊断的资料,模拟专家诊断思维,分析故障形成原因,按照树枝状逐级细化的方式,建立变压器故障树。然后针对故障树分析过程复杂,计算量大的问题,化简故障树的分析步骤。利用变压器的状态量,按故障树的结构分层进行匹配,找到对应的故障节点。运行人员参考系统诊断结果,结合现场的实际情况、统计资料和专家诊断结果等信息,对故障进行处理。最后,通过实例证明了本文方法的有效性和可行性。

A transformer fault diagnostic method based on multi-parameter fusion fault tree is presented in the paper. Firstly, the fault diagnostic data of transformer is collected. The fault tree is created through simulating the expert thinking and analyzing the fault reason. Furthermore, to solve the problem of the complicated analysis process and large calculation, the steps of fault tree are reduced in the paper. The fault tree nodes are matched using the state values of transformer, and then the fault is found. Referring to the diagnosis result in the paper, the operators deal with the fault combined with actual situation, statistics data and expert advice. Finally, the feasibility and effectiveness of this developed method is illustrated by an example of transformer fault diagnosis.