Partial Discharge Pattern Recognition of High-voltage Cables Based on Random Forest Method

LIU Wenhao; WU Yijiang; LI Wenze; WANG Hongyu; Ashfaque Ahmed Bhatti; PENG Xiaosheng; HE Shunji

doi:10.13296/j.1001-1609.hva.2022.06.023

您当前的位置：

首页 >

文章列表页 >

Partial Discharge Pattern Recognition of High-voltage Cables Based on Random Forest Method

更新时间：2025-04-26

- Partial Discharge Pattern Recognition of High-voltage Cables Based on Random Forest Method
- High VoltageApparatus Vol. 58, Issue 6, Pages: 165-170+177(2022)
- 作者机构：
  
  1. 广东电网公司中山供电局
  2. 华中科技大学电气与电子工程学院
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2022.06.023
  CLC： TM855
- Published：2022
- 稿件说明：
移动端阅览
LIU Wenhao, WU Yijiang, LI Wenze, et al. Partial Discharge Pattern Recognition of High-voltage Cables Based on Random Forest Method[J]. High voltageapparatus, 2022, 58(6): 165-170+177.
DOI：

LIU Wenhao, WU Yijiang, LI Wenze, et al. Partial Discharge Pattern Recognition of High-voltage Cables Based on Random Forest Method[J]. High voltageapparatus, 2022, 58(6): 165-170+177. DOI： 10.13296/j.1001-1609.hva.2022.06.023.

摘要

不同类型的高压电缆局部放电（简称局放）模式识别是该领域的难题。部分电缆局放类型之间相似度较大，识别困难。针对该问题，文中提出了一种基于随机决策森林（RF）的高压电缆局放模式识别方法。首先，制作了5种高压电缆人工缺陷，结合IEC 60270—2015系统和高频电流互感器（HFCT）进行实验，获取局放数据，并进行局放特征提取。其次，介绍了随机森林算法的原理和基于随机森林的高压电缆局放模式识别流程。最后，基于实验所得数据开展了局放模式识别方法有效性验证，确定了随机森林每个节点处特征子集中特征个数、节点分裂规则、树的棵数3个参数，并与传统的决策树、BP神经网络和支持向量机（SVM）3种方法进行比较。结果显示，与上述3种方法相比，随机森林算法对高相似度局放识别能力更强。

Abstract

Partial discharge（PD）pattern recognition of different types of high voltage cables is the difficulty of the field. The similarity among PD types of partial cable is large and difficult to be recognized. In view of this problem

a kind of high voltage cables PD pattern recognition method based on random forest（RF）is proposed in the paper.First

the artificial defects of five types of high voltage cables are made and the test is performed in combination with IEC 60270—2015 system and high frequency current transformer（HFCT）so to obtain PD data and extract the PD feature. Then

the principle of random forest algorithm and pattern recognition of high voltage cable based on random forest are introduced. Finally

the effectiveness verification of the PD mode recognition is developed on the basis of the data obtained by the experiment and such three parameters as the number of features in the feature subset at each node

the node splitting rule and the number of trees are determined. And the pattern recognition results based on the random forest method is compared with such three methods as decision tree

BPNN and SVM. The results show that the random forest algorithm

compared with the above three kinds of methods

has a stronger recognition ability for high similarity PD recognition.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

PARTIAL DISCHARGE PATTERN RECOGNITION FOR XLPE POWER CABLE BASED ON BACK-PROPAGATION ARTIFICIAL NEURAL NETWORK

Pattern Recognition of XLPE Cable Partial Discharge Base on Gray Image

Pattern Recognition of Partial Discharge in Oil-paper Insulation Based on Analytic Hierarchy Process

Polar Spectrum of Partial Discharge and Pattern Recognition

Research on a Cascade Partial Discharge Locating Method for High Voltage Switchgear Cabinet

Related Author

YANG Xiao-hua 1

LIAO Rui-jin 1

HU Jian-lin 1

BAI Shi-xiong 2

LIAO Rui-jin

YOU Deng-liang

ZHOU Quan

LIU Ling

Related Institution

Chongqing University

Nanchong Electrical Power Bureau,Nanchong

四川南充637000

The Key Laboratory of High Voltage Engineering and Electrical New Technology,Ministry of Education,Chongqing University

重庆大学高电压与电工新技术教育部重点实验室重庆 400044

⁰