基于空间分析理论改进的变压器Duval Pentagon1故障诊断方法  被引量:3

Improved Duval Pentagon1 Fault Diagnosis Method for Transformer Based on Spatial Analysis Theory

在线阅读下载全文

作  者:张丞鸣 谢菊芳[1] 余松 唐超[1,2] 胡东[1,2] ZHANG Chengming;XIE Jufang;YU Song;TANG Chao;HU Dong(College of Engineering and Technology,Southwest University,Chongqing 400715,China;International R&D Center for New Technologies of Smart Grid and Equipment,Southwest University,Chongqing 400715,China)

机构地区:[1]西南大学工程技术学院,重庆400715 [2]西南大学智能电网及装备新技术国际研发中心,重庆400715

出  处:《高电压技术》2022年第6期2255-2264,共10页High Voltage Engineering

基  金:国家自然科学基金(51977179)。

摘  要:为了解决Duval Pentagon1法不能表征各种变压器故障模式可信度的问题,提出了一种基于空间分析理论改进的方法。首先,采用核密度估计(kernel density estimation,KDE)方法对各种变压器故障模式数据进行空间分布密度分析;其次,应用B样条理论构造各种变压器故障模式密度曲面;最后,利用空间叠置分析方法实现变压器故障模式识别。实例分析结果表明:与Duval Pentagon1法相比,改进后的Duval Pentagon1法的总体诊断准确率提升了8.42%,并且能够定量地表征各种变压器故障模式可信度。研究结果可为变压器油中溶解气体图形分析方法的改进提供参考。To solve the problem that the Duval Pentagon1 method cannot represent the credibility of various transformer fault modes,an improved method based on the spatial analysis theory is proposed in this paper.Firstly,the kernel density estimation(KDE)method is used to analyze the spatial distribution density of various transformer fault mode data.Secondly,the B-spline theory is applied to construct the density surfaces of various transformer fault modes.Finally,the spatial overlay analysis method is used to identify the transformer fault modes.The case analysis results show that the overall diagnostic accuracy of the improved Duval Pentagon1 method is 8.42%higher than that of the Duval pentagon1method,and it can quantitatively characterize the credibility of each transformer fault mode.The research results in this paper can provide reference for improvement of the graphic analysis method of dissolved gas in transformer oil.

关 键 词:变压器 故障诊断 Duval Pentagon1法 油中溶解气体分析 空间分析理论 

分 类 号:TM411[电气工程—电器]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象