油浸式变压器故障率模型及故障诊断研究

发布时间：2018-01-11 00:20

本文关键词：油浸式变压器故障率模型及故障诊断研究　出处：《浙江大学》2014年硕士论文　论文类型：学位论文

【摘要】：电力变压器是电网中的核心部件,也是电网公司的重要资产,其安全稳定运行意义重大。在实时监测变压器状态、运行条件的基础上,可对变压器的故障率进行评价,进而合理安排检修维护措施来降低设备发生故障的风险。此外,对已经故障停运或故障率高的变压器进行故障诊断,迅速找到故障原因并采取相应修复措施,能够有效减少维护时间,降低变压器停运造成的经济损失。因此,本文针对油浸变压器的故障率建模方法和故障诊断方法,开展了以下研究。基于比例故障率模型(Proportional Hazard Model, PHM)和油中溶解气体信息提出了一种综合考虑老化和设备状态的油浸变压器故障率模型。本文提出的模型中,比例故障率模型的基准故障率函数采用常用的温升老化模型,连接函数中的协变量选择了能够全面客观反映设备状况的油中溶解气体信息,然后推导了故障前时间的概率密度分布,并给出了使用极大似然估计拟合参数的方法。通过算例证明了提出模型的正确性。支持向量机(Support Vector Machine, SVM)可用于变压器故障诊断,针对现有SVM方法在样本故障特征不明显情况下有误分类的情况,提出了一种基于支持向量机多分类概率输出的变压器故障诊断方法,此方法可以得到发生不同类型故障的可能性,即故障类别的概率,经过进一步分析后给出诊断结论。算例表明本方法在继承了SVM方法优点的基础上,提供了概率信息,对现有SVM方法误诊断样本也能给出可能存在的故障,弥补了现有SVM方法在变压器故障特征不明显条件下的不足。在电网公司“调控一体化”的大背景下,进一步挖掘分析设备监测信息,为给调控中心设定潜在事故预警预案、制定运行方式、合理安排检修计划、优化调度策略提供基础,实验室项目组开发了“基于输变电设备可载性分析的智能电网风险评估与决策系统”。本文介绍了此系统的功能,然后重点介绍和展示了笔者开发的输变电设备健康评估功能中变压器状态评估和故障诊断模块、基于SVG的信息展现模块。
[Abstract]:Power transformer is the core part of the power network, and also an important asset of the power grid company. Its safe and stable operation is of great significance. On the basis of real-time monitoring of transformer status and operating conditions. The failure rate of transformer can be evaluated, and maintenance measures can be arranged reasonably to reduce the risk of equipment failure. In addition, fault diagnosis can be carried out for transformers that have been out of service or have high failure rate. Finding the fault cause quickly and taking the corresponding repair measures can effectively reduce the maintenance time and reduce the economic loss caused by the transformer outage. In this paper, the fault rate modeling method and fault diagnosis method of oil-immersed transformer are studied as follows. Proportional Hazard Model based on proportional failure rate model. PHM) and dissolved gas information in oil a failure rate model for oil-immersed transformers considering aging and equipment state is proposed. The benchmark failure rate function of the proportional failure rate model adopts the commonly used temperature rise aging model and the covariable in the connection function selects the dissolved gas information in the oil which can reflect the equipment condition comprehensively and objectively. Then, the probability density distribution of time before failure is deduced, and the method of using maximum likelihood estimation to estimate fitting parameters is given, and the correctness of the proposed model is proved by an example. Support Vector Machine support Vector Machine can be used in transformer fault diagnosis. In this paper, a transformer fault diagnosis method based on multi-classification probability output of support vector machine (SVM) is proposed to solve the problem that the existing SVM method has wrong classification under the condition that the sample fault feature is not obvious. The probability of different types of faults, that is, the probability of fault types, can be obtained by this method. After further analysis, a diagnosis conclusion is given. The example shows that this method inherits the advantages of SVM method. The probabilistic information is provided, and the possible faults can be given for the existing SVM method, which can make up for the deficiency of the existing SVM method under the condition that the fault characteristics of the transformer are not obvious. Under the background of "the integration of regulation and control" of the power grid company, the monitoring information of the equipment is further excavated and analyzed, the potential accident warning plan is set up for the control center, the operation mode is worked out, and the maintenance plan is arranged reasonably. The laboratory project team has developed a smart grid risk assessment and decision system based on load analysis of power transmission and transformation equipment. The functions of the system are introduced in this paper. Then the transformer condition assessment and fault diagnosis module and the information display module based on SVG are introduced and displayed in the health evaluation function of transmission and transformer equipment developed by the author.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM411

【参考文献】