摘要
变压器作为电力系统的关键设备,维护其安全稳定运行具有重要的意义。由于变压器自身结构复杂,利用单一信息的传统故障诊断方法对其进行诊断具有一定的局限性。结合变压器油气数据,利用数据融合原理,将BP神经网络和证据理论相结合,设计了多源信息融合的变压器故障诊断模型,并利用现场数据对该模型进行测试。测试结果表明,该模型能有效地进行变压器的故障诊断,与传统方法相比提高了故障诊断的正确率,具有较高的理论意义和应用价值。
As the key equipment of the electric power system, the power transformer plays an important role in maintaining system＇s stability. Due to the transformer＇s complicated structure, there is limitation in traditional diag-nosis methods based on single information. In this paper, a synthetic diagnosis approach using BP neural network and evidence theory for transformer fault diagnosis is presented. combies multi-sensor （ dissolved gas analysis ） DGA data. Field data is used to test the proposed approach. Results demonstrate that the presented method can ef-fectively conduct fault diagnosis of power transformers. Compared with the traditional method, the new approach improves the accuracy and has high theoretical significance and application value.
出处
《电力科学与工程》
2013年第11期21-26,共6页
Electric Power Science and Engineering
关键词
变压器
故障诊断
多源信息融合
BP神经网络
油中溶解气体
power transformer
fault diagnosis
multi-source information fusion
BP neural network
dissolved gasanalysis
