基于能量加权DGA的变压器潜伏性故障诊断及故障率估计方法
发布时间：2018-10-15 12:39
【摘要】：电力变压器是电力系统中重要的输变电设备之一,在电力系统中处于枢纽地位,其运行的安全可靠性直接关系到电力系统的安全。对变压器进行准确的故障诊断,把握变压器的状态,对于保证其安全稳定运行有着重要的意义。变压器潜伏性故障发展时间较长,目前比较有效的检测手段是油中溶解气体分析技术(Dissolved Gas Analysis,DGA),可以发现变压器是否发现故障,区分故障类别,并且对故障的严重程度进行判断。此外,对变压器的潜伏性故障故障率做出准确的估计,对于指导变压器的运维等生产活动也有着重要的意义。基于以上背景,本文主要围绕变压器故障诊断和潜伏性故障的故障率估计方法展开研究。本文首先分析变压器故障类型以及变压器故障时油中溶解气体的特征,并对目前应用于电力设备故障诊断的人工智能方法及其优缺点进行综述;根据极限学习机方法的特点及优势,建立了基于极限学习机的故障分类模型,通过算例验证该模型在故障诊断方面的有效性,为基于DGA的变压器故障诊断提供一种新思路。然后,本文在热动力学中焓的概念基础上,通过对变压器油分解化学反应进行分析,根据标准生成焓对不同故障气体进行加权,引入能量加权DGA(Energy Weighted DGA,EWDGA)概念,分析其在变压器故障严重程度判断方面的应用,并利用算例说明EWDGA在判别故障程度时比传统的产气率数据有着更客观的判断。最后,本文基于EWDGA和马尔可夫过程,提出计及能量加权产气率数据的变压器潜伏性故障的故障率估计方法,并通过算例说明本文提出的方法与传统方法相比,能够更好的区分产气率相同的状态下,变压器不同故障发展程度时的潜伏性故障的故障率。
[Abstract]:Power transformer is one of the important transmission and transformation equipment in power system. It is in the pivotal position in the power system. The safety and reliability of its operation is directly related to the security of power system. It is of great significance to make accurate fault diagnosis and grasp the state of transformer to ensure its safe and stable operation. The latent fault of transformer has a long development time. At present, the more effective detection method is dissolved gas analysis technology (Dissolved Gas Analysis,DGA) in oil. It can find out whether the transformer finds fault, distinguish the type of fault, and judge the severity of fault. In addition, it is of great significance to estimate the latent failure rate of transformers for guiding the operation and maintenance of transformers. Based on the above background, this paper focuses on transformer fault diagnosis and fault rate estimation of latent faults. This paper first analyzes the types of transformer faults and the characteristics of dissolved gases in oil when transformer faults occur, and summarizes the current artificial intelligence methods used in fault diagnosis of power equipment and their advantages and disadvantages. According to the characteristics and advantages of extreme learning machine, a fault classification model based on ultimate learning machine is established. The validity of the model in fault diagnosis is verified by an example, which provides a new way for transformer fault diagnosis based on DGA. Then, based on the concept of enthalpy in thermodynamics, by analyzing the decomposition chemical reaction of transformer oil, according to the standard enthalpy of formation, the concept of energy weighted DGA (Energy Weighted DGA,EWDGA is introduced. The application of EWDGA in judging the fault severity of transformer is analyzed, and an example is given to show that EWDGA has a more objective judgment than the traditional gas production rate data in judging the fault degree. Finally, based on EWDGA and Markov process, a fault rate estimation method for transformer latent faults considering energy-weighted gas production rate data is proposed, and an example is given to illustrate the comparison between the proposed method and the traditional method. It can better distinguish the failure rate of latent faults under the condition of the same gas production rate and different development degree of transformer faults.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM407
本文编号：2272582
[Abstract]:Power transformer is one of the important transmission and transformation equipment in power system. It is in the pivotal position in the power system. The safety and reliability of its operation is directly related to the security of power system. It is of great significance to make accurate fault diagnosis and grasp the state of transformer to ensure its safe and stable operation. The latent fault of transformer has a long development time. At present, the more effective detection method is dissolved gas analysis technology (Dissolved Gas Analysis,DGA) in oil. It can find out whether the transformer finds fault, distinguish the type of fault, and judge the severity of fault. In addition, it is of great significance to estimate the latent failure rate of transformers for guiding the operation and maintenance of transformers. Based on the above background, this paper focuses on transformer fault diagnosis and fault rate estimation of latent faults. This paper first analyzes the types of transformer faults and the characteristics of dissolved gases in oil when transformer faults occur, and summarizes the current artificial intelligence methods used in fault diagnosis of power equipment and their advantages and disadvantages. According to the characteristics and advantages of extreme learning machine, a fault classification model based on ultimate learning machine is established. The validity of the model in fault diagnosis is verified by an example, which provides a new way for transformer fault diagnosis based on DGA. Then, based on the concept of enthalpy in thermodynamics, by analyzing the decomposition chemical reaction of transformer oil, according to the standard enthalpy of formation, the concept of energy weighted DGA (Energy Weighted DGA,EWDGA is introduced. The application of EWDGA in judging the fault severity of transformer is analyzed, and an example is given to show that EWDGA has a more objective judgment than the traditional gas production rate data in judging the fault degree. Finally, based on EWDGA and Markov process, a fault rate estimation method for transformer latent faults considering energy-weighted gas production rate data is proposed, and an example is given to illustrate the comparison between the proposed method and the traditional method. It can better distinguish the failure rate of latent faults under the condition of the same gas production rate and different development degree of transformer faults.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM407
【相似文献】
相关期刊论文 前10条
1 唐铁英;徐建文;;电容式电压互感器潜伏性故障发现及原因分析[J];中国电力教育;2011年18期
2 陈超宗;时光;余维祥;李桂兰;唐赤峰;;用液相色谱仪判断充油电力设备潜伏性故障的初步探讨[J];分析仪器;1982年02期
3 孙平;王志明;;浅议用气相色谱法分析潜伏性故障[J];华北电力技术;1992年02期
4 黎笑兴;气相色谱法在检测变压器潜伏性故障中的应用[J];广东电力;1996年02期
5 孙丽华;利用色谱分析判断设备内部潜伏性故障[J];黑龙江电力技术;1998年05期
6 周志敏;色谱分析在变压器潜伏性故障诊断中的应用[J];大众用电;2003年11期
7 赵铭刚;;色谱分析在充油电气设备内部潜伏性故障诊断中的应用[J];中国新技术新产品;2010年02期
8 倪健;变压器内部潜伏性故障的诊断[J];农村电气化;2001年05期
9 刘艳苹;;气相色谱应用于充油电气设备内部潜伏性故障的诊断[J];内蒙古石油化工;2011年12期
10 宋洁;杜正旺;;主变压器潜伏性故障的检测与分析[J];电气制造;2013年08期
相关会议论文 前1条
1 张瑞春;姜吉平;;用气相色谱法检测变压器潜伏性故障几种判断方法的分析[A];华东六省一市电机工程（电力）学会输配电技术研讨会2002年年会山东电机工程学会交流论文集[C];2002年
相关硕士学位论文 前1条
1 袁锦浩;基于能量加权DGA的变压器潜伏性故障诊断及故障率估计方法[D];上海交通大学;2015年
，本文编号：2272582
本文链接：https://www.wllwen.com/kejilunwen/dianlilw/2272582.html
