基于RBF神经网络的GIS设备特高频局放检测技术的研究  

Research on Ultra High Frequency Partial Discharge Detection Technology for GIS Equipment Based on RBF Neural Network

在线阅读下载全文

作  者:张一博 Zhang Yibo(State Grid Shanxi Electric Power Company Ultra High Voltage Substation Branch,Taiyuan Shanxi 030000,China)

机构地区:[1]国网山西省电力公司超高压变电分公司,山西太原030000

出  处:《现代工业经济和信息化》2024年第12期66-68,73,共4页Modern Industrial Economy and Informationization

摘  要:为控制局放检测结果的偏差,基于RBF神经网络的应用,以GIS设备为例,开展特高频局放检测技术的研究。根据安装方式的不同,将传感器分为内置式与外置式,通过布置传感器,实现对GIS设备运行信号的采样;对采样的数据进行归一化处理,将处理后的数据录入RBF神经网络输入层,接收的数据采用径向基层的高斯函数作为激活函数,利用激活函数,对数据进行训练,实现对采样信号的处理与特征提取;计算两个传感器接收到的信号之间的相似度,以此为依据,实现对局放信号的时差定位与检测。对比实验表明:应用设计的方法进行GIS设备特高频局放检测,检测结果幅值偏差在±0.2 dBm V范围内,偏差相对较低。In order to control the deviation of partial discharge detection results,based on the application of RBF neural network and taking GIS equipment as an example,research on ultra-high frequency partial discharge detection technology is carried out.According to the different installation methods,sensors are divided into built-in and external types.By arranging sensors,the sampling of GIS equipment operation signals can be achieved;Normalize the sampled data,input the processed data into the input layer of the RBF neural network,and use the Gaussian function of the radial base layer as the activation function to train the data and achieve processing and feature extraction of the sampled signal;Calculate the similarity between the signals received by two sensors,and based on this,achieve time difference localization and detection of partial discharge signals.Comparative experiments show that using the designed method for ultra-high frequency partial discharge detection in GIS equipment results in amplitude deviation within ± 0.2 dBmV,which is relatively low.

关 键 词:RBF神经网络 特征提取 时差定位 局放检测 特高频 GIS设备 

分 类 号:TM643[电气工程—电力系统及自动化]

 

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