故障诊断论文：变压器故障多信息融合诊断方法研究.docVIP

故障诊断论文：变压器故障多信息融合诊断方法研究 【中文摘要】变压器是电力系统的枢纽设备,在电力系统输变电过程中起着举足轻重的作用,其运行的可靠性直接关系着电力系统的安全运行。为了保证电力系统的安全稳定运行,必须加强对变压器的故障诊断。目前,用于变压器故障诊断研究最多的特征量是油中溶解气体含量和局部放电信号。油中溶解气体分析能够诊断变压器故障类型,局部放电的模式识别能够及时发现绝缘内部局部缺陷及放电发展程度,可以将故障消灭在萌芽状态。因此本文利用油气特征含量和局部放电信号进行变压器综合故障诊断。本文首先基于油中溶解气体特征含量构建BP网络结构,采集数据样本,进行网络训练,然后应用信息融合原理搭建分层故障诊断模型,分析判断变压器的故障类型,并在此基础上应用DS证据理论缩小故障范围。论文针对变压器放电故障的放电类型进行诊断,在实验室搭建了三种典型缺陷局部放电模型,采用升压法对不同放电模型进行试验,分析了各种不同放电模型的发展特性,提取了统计特征量,为局部放电模型识别提供输入参量。同时在实验过程中,采集了油气数据来研究变压器典型缺陷局部放电发展过程中的油中溶解气体变化规律,发现不同缺陷的局部放电,油中溶解气体的变化规律不同,可以应用油中溶解气体进行局部放电类型的诊断。通过以上的研究,本文提出了一种基于油气含量和局部放电信号统计特征参量的多信息融合变压器故障诊断的研究方法,经过对局部放电模型识别的实例的验证,证明该方法可行;最后,通过Labview这个软件平台开发了一种分层、可靠、开放的变压器多信息融合故障诊断系统。 【英文摘要】As the core apparatus in electrical power system, power transformer plays a decisive role in the transmission and distribution of power system. Its operating reliability attaches importance to system抯 safety directly. In order to insure the stability of power grid, it is indispensable to be reinforced the transformer fault diagnosis. Recently, gas-oil and part-discharging signals are most frequently studied in transformer fault diagnosis. Dissolved gas analysis in oil is able to diagnose transformer faults, partial discharge pattern recognition can discover internal local defects and hidden danger of failures of transformer, which can fault nipped in the bud. In this back ground, this paper gives a study of transformer synthetic diagnosis with oil gas feature content and partial discharge signals.Based on the oil gas feature content, this paper contracts the BP neural network and acquisites the data sample to train the neural network. Then a layered fault diagnosis model was established with information fusion theory. This method can diagnosis the transformer fault type, and find the fault location on future.After analyze the common failure of partial discharge, this paper designes three kinds of typical defects model of partial discharge, and carries on the corresponding discharge test with step-up method