Fault Diagnosis Method of Power Transformers Based on Support Vector Machine and Kriging Model

LIU Qiang; DING Yu

doi:10.13296/j.1001-1609.hva.2018.12.042

您当前的位置：

首页 >

文章列表页 >

Fault Diagnosis Method of Power Transformers Based on Support Vector Machine and Kriging Model

更新时间：2025-04-26

- Fault Diagnosis Method of Power Transformers Based on Support Vector Machine and Kriging Model
- High VoltageApparatus Vol. 54, Issue 12, Pages: 274-280(2018)
- 作者机构：
  
  辽宁石油化工大学信息与控制工程学院
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2018.12.042
  CLC： TM407
- Published：2018
- 稿件说明：
移动端阅览
[1]柳强,丁宇.基于SVM和Kriging模型的变压器故障诊断方法[J].高压电器,2018,54(12):274-280.

LIU Qiang, DING Yu. Fault Diagnosis Method of Power Transformers Based on Support Vector Machine and Kriging Model[J]. High voltageapparatus, 2018, 54(12): 274-280.
[1]柳强,丁宇.基于SVM和Kriging模型的变压器故障诊断方法[J].高压电器,2018,54(12):274-280. DOI： 10.13296/j.1001-1609.hva.2018.12.042.

LIU Qiang, DING Yu. Fault Diagnosis Method of Power Transformers Based on Support Vector Machine and Kriging Model[J]. High voltageapparatus, 2018, 54(12): 274-280. DOI： 10.13296/j.1001-1609.hva.2018.12.042.

摘要

为了保证变压器供电的可靠性

结合油中溶解气体分析法（DGA）

提出了基于支持向量机（SVM）和Kriging模型的双级分类算法

实现变压器的潜伏性故障的预测。首先

以SVM作为初级分类器

将待测DGA数据分类为正常、放电故障或过热故障三种状态;其次

以Kriging模型作为次级分类器

对待测数据进行二次分类

完成精确诊断。为提高算法精度

应用改进的粒子群优化算法分别对SVM的核函数参数、惩罚系数以及Kriging模型的相关参数进行优化。所提方法集成了粒子群优化算法的快速寻优能力

SVM的分类能力与Kriging模型高精度拟合等优点

因此有效地提高了诊断精度。最后通过变压器故障诊断实例验证了文中方法的有效性。

Abstract

In order to ensure the reliability of transformer power supply

a fault diagnosis modeling method based on Kriging model and support vector machine（SVM）is proposed

which is applied to the transformer fault diagnosis with analysis of dissolved gases in oil.Firstly

SVM as the first classifier to diagnosis the DGA data to be measured as normal

discharge fault or overheating fault three states.Subsequently

Kriging model as the second classifier to diagnose the measured data again.To improve accuracy of the proposed method

the modified particle swarm optimization algorithm is used to optimize the kernel function parameters and the penalty coefficient of SVM and the related parameters of the Kriging model.The proposed method which integrates the fast optimization ability of particle swarm optimization algorithm is based on the classification ability of SVM and the high precision fitting of Kriging model.Therefore

the modeling accuracy is improved effectively.Finally

the validity of this method is verified by several examples of transformer fault diagnosis.

关键词

Keywords

references

Views

223

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Fault Diagnosis Model of Transformer Based on GJO Feature Optimization and AO-RF

Research on Diagnosis Method of Transformer Winding Deformation Fault Based on Dynamic Time Warping

New Approach to Transformer Fault Diagnosis Based on ANN

APPLICATION OF INFORMATION FUSION TECHNIQUE TO TRANSFORMER FAULT DIAGNOSIS

Research on Power Transformer Fault Diagnosis with BPNN Method

Related Author

YE Yulin

LIU Sen

HAN Xiaohui

DU Zhenbin

LI Bin

LYU Jie

XUE Yang

ZHAO Chunlin

Related Institution

China Nuclear Power Engineering Co., Ltd.

School of Electrical Engineering, Hebei University of Science and Technology

Hebei Provincial Key Laboratory of Electromagnetic & Structural Performance of Power Transmission and Transformation Equipment, Baoding Tianwei Baobian Electric Co., Ltd.

Baoding Tianweixinyu Technology Development Co., Ltd.

Hebei Weixun Electric Power Automation Equipment Co., Ltd.

⁰