D-S Multi Information Fusion GIL Partial Discharge Joint Recognition Method Based on Neural Network

被引：0
|
作者：
Ai J. [1 ]
Niu H. [1 ]
Chen Z. [1 ]
Li X. [1 ]
Zeng Y. [1 ]
Zhuang X. [2 ]
机构：
[1] School of Electric Power, South China University of Technology, Guangzhou
[2] China Southern Power Grid EHV Transmission Company, Guangzhou
来源：
Gaodianya Jishu/High Voltage Engineering | 2022年 / 48卷 / 12期
关键词：
D-S multi-information fusion; gas insulated transmission line; joint identification; neural network; partial discharge;
D O I：
10.13336/j.1003-6520.hve.20211249
中图分类号：
学科分类号：
摘要：
In order to improve the recognition rate of partial discharge fault of typical defects in GIL (gas insulated transmission line). In this paper, a joint partial discharge recognition method of GIL typical defects based on D-S multi information fusion of neural network is proposed. With the partial discharge’s ultrasonic signal, UHF signal, and acoustic-electric (ultrasound-ultra-high frequency) of GIL typical defect, the method construct their corresponding Hankel matrix and extract their singular value characteristics, utilize BP neural network to identify the infect type. Put this preliminary recognition result as the evidence body to calculate the reliability distribution based on the DS evidence synthesis rule, and finally identify the infect type on base of the decision rule. The research shows that by the BP neural network recognition method, the total recognition rates of typical defect in GIL applying partial discharge’s UHF signals, ultrasonic signals and combined acoustic-electrical signals are 85%, 75%, and 97% respectively. Furthermore the overall recognition rate can be improved to over 99% resorting to DS multi-information fusion. © 2022 Science Press. All rights reserved.
引用
收藏
页码：4925 / 4932
页数：7
相关论文
共 20 条
  • [1] XU Wenjia, XIANG Zutao, MA Qiyan, Et al., On-line short circuit fault location method for long distance GIL based on current transformer, Electrical Measurement & Instrumentation, 56, 21, pp. 122-128, (2019)
  • [2] QI Bo, ZHANG Guixin, LI Chengrong, Et al., Research status and prospect of gas-insulated metal enclosed transmission line, High Voltage Engineering, 41, 5, pp. 1466-1473, (2015)
  • [3] ZHAO Jun, CHEN Weijiang, BIAN Kai, Et al., Distribution characteristics of spatial magnetic field in UHV GIL gallery in short-circuit fault condition, High Voltage Engineering, 47, 5, pp. 1866-1871, (2021)
  • [4] WANG Jingfei, ZHANG Qiang, LI Xiangbin, Et al., Fault analysis and improvement measures of GIL three-pillar insulator in UHVDC transmission project, High Voltage Apparatus, 56, 1, pp. 246-252, (2020)
  • [5] TANG Ju, LI Wei, OUYANG Youpeng, Partial discharge pattern recognition using discrete wavelet transform and singular value decomposition, High Voltage Engineering, 36, 7, pp. 1686-1691, (2010)
  • [6] GUO Moufa, XU Lilan, MIAO Xiren, Et al., A vibration signal feature extraction method for distribution switches based on singular value decomposition of time-frequency matrix, Proceedings of the CSEE, 34, 28, pp. 4990-4997, (2014)
  • [7] YANG Qi, ZENG Huarong, HUANG Huan, Et al., Study on predischarge recognition of hidden dangers in transmission lines based on BP neural network, Power Systems and Big Data, 23, 3, pp. 47-54, (2020)
  • [8] ZHOU Sha, JING Liang, Pattern recognition of partial discharge based on moment features and probabilistic neural network, Power System Protection and Control, 44, 3, pp. 98-102, (2016)
  • [9] LU Qiwei, WANG Tao, LI Zongrui, Et al., Detection method of series arcing fault based on wavelet transform and singular value decomposition, Transactions of China Electrotechnical Society, 32, 17, pp. 208-217, (2017)
  • [10] HUANG Liang, TANG Ju, LING Chao, Et al., Pattern recognition for partial discharge based on multi-feature fusion technology, High Voltage Engineering, 41, 3, pp. 947-955, (2015)
12