Application of Support Vector Machine in Breakdown Voltage Prediction of Complex Air Gap

TANG Liezheng; RUAN Jiangjun; QIU Zhibin; HUANG Daochun; LIAO Yifan; HUANG Congpeng

doi:10.13296/j.1001-1609.hva.2017.12.029

您当前的位置：

首页 >

文章列表页 >

Application of Support Vector Machine in Breakdown Voltage Prediction of Complex Air Gap

更新时间：2025-04-26

- Application of Support Vector Machine in Breakdown Voltage Prediction of Complex Air Gap
- High VoltageApparatus Vol. 53, Issue 12, Pages: 175-180+187(2017)
- 作者机构：
  
  1. 武汉大学电气工程学院
  2. 南方电网科学研究院
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2017.12.029
  CLC： TM85
- Published：2017
- 稿件说明：
移动端阅览
[1]唐烈峥,阮江军,邱志斌,黄道春,廖一帆,黄从鹏.支持向量机在组合空气间隙击穿电压预测中的应用[J].高压电器,2017,53(12):175-180+187.

TANG Liezheng, RUAN Jiangjun, QIU Zhibin, et al. Application of Support Vector Machine in Breakdown Voltage Prediction of Complex Air Gap[J]. High voltageapparatus, 2017, 53(12): 175-180+187.
[1]唐烈峥,阮江军,邱志斌,黄道春,廖一帆,黄从鹏.支持向量机在组合空气间隙击穿电压预测中的应用[J].高压电器,2017,53(12):175-180+187. DOI： 10.13296/j.1001-1609.hva.2017.12.029.

TANG Liezheng, RUAN Jiangjun, QIU Zhibin, et al. Application of Support Vector Machine in Breakdown Voltage Prediction of Complex Air Gap[J]. High voltageapparatus, 2017, 53(12): 175-180+187. DOI： 10.13296/j.1001-1609.hva.2017.12.029.

摘要

为获取含悬浮导体的组合空气间隙击穿特性

建立了基于支持向量机和电场特征量的组合空气间隙击穿电压预测模型

通过电场特征量表征间隙结构

采用支持向量机建立电场特征量与击穿电压的关联性

以少量典型间隙试验数据作为训练样本

对棒—板—球、棒—板—棒以及球—板—球组合间隙的工频击穿电压进行了预测

通过对比击穿电压试验数据

3种间隙的击穿电压预测结果平均绝对百分比误差分别为4.5%、3.1%和2.8%

最大相对误差均在9%以内

表明所提方法具有较高精度

为组合间隙击穿电压的预测提供了新途径。

Abstract

To obtain breakdown characteristics of complex air gap with floating metals

a prediction model for the breakdown voltages of the complex air gap was established based on support vector machine and electric field features.The gap structure was characterized by electric field features and the relationship between the breakdown voltage

and the electric field features was constructed by support vector machine.Taking some typical gaps as the training sample data

the breakdown voltages of rod-plane-sphere gaps

rod-plane-rod gaps and sphere-plane-sphere gaps were predicted.By comparing with the test data

the mean absolute percentage errors of the predicted breakdown voltage of the three kinds of gaps are 4.5%

3.1% and 2.8%

respectively

and the maximum relative error is within 9%

showing a high accuracy of the prediction method.This study may provide a new approach to predict the complex air gap breakdown voltage.

关键词

Keywords

references

Views

105

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Prediction of Breakdown Voltage of Transformer Oil via Multi-parameter Correlation Based on BP Neural Network

Review on Prediction Methods for Breakdown Voltage of Air Gap

Research on GIL Mechanical Fault Detection System Based on Combined Vibration and Acoustic Location Technology

Support Vector Machine Load Modeling Parameter Identification Method Based on Leave-one-out Cross Validation Method

Green Digital Management and Early Warning Method for Power Grid Infrastructure Files Based on Improved PKCNN

Related Author

LI Rui

CAO Shun-an

SHENG Kai

SHU Shengwen

LIU Chang

RUAN Jiangjun

XU Aihua

ZANG Chunyan

Related Institution

College of Power and Machine,Wuhan University

State Grid Fujian Electric Power Co.,Ltd.,Electric Power Research Institute

State Grid Fujian Electric Power Co.,Ltd.,Fuzhou Power Supply Company

Hubei Institute of Measurement and Testing Technology

School of Electrical and Electronic Engineering, Huazhong University of Science and Technology

⁰