摘要
为了克服了人工神经网络(ANN)中存在的过拟合、收敛速度慢、容易陷入局部极值等缺点,提出了基于粒子群优化支持向量机(PSO-SVM)的变压器故障诊断方法,即将粒子群优化算法(PSO)用于SVM参数优化。PSO是一种智能群体搜索方法,它源于对鸟类捕食行为的研究。这种方法不仅具有很强的全局搜索能力,而且容易实现,适合于SVM参数优化。变压器故障诊断实例分析结果证明,PSO-SVM的诊断精度高于IEC三比值法、BP神经网络、普通的SVM,PSO-SVM适用于电力变压器故障诊断。
In order to solve the problem of over-fitting, local optimal solution and low convergence rate existed in artificial neural networks（ANN）, the proposed PSO-SVM model is applied to fault diagnosis of power transformer in the paper. Among which particle swarm optimization（PSO） is used to determine free parameters of support vector machine. PSO is an intelligent swarm optimization method, which derives from the research for behaviour of bird flocking. The method not only has strong global search capability, but also is very easy to implement. Thus, PSO is suitahle to determine free parameters of support vector machine（SVM）. Finally, fault diagnosis examples are used to illustrate the performance of proposed PSO-SVM model. The experimental results indicate that the PSO-SVM method can achieve higher diagnostic accuracy than IEC three ratios, normal SVM classifier, artificial neural network. Consequently, the PSO SVM model is a proper alternative for fault diagnosis of power transformer.
出处
《高电压技术》
EI
CAS
CSCD
北大核心
2009年第3期509-513,共5页
High Voltage Engineering
基金
国家高技术研究发展计划(863)(2007AA10Z209
2007AA04Z434)~~
关键词
故障诊断
粒子群优化
支持向量机
电力变压器
参数优化
分类算法
统计学习理论
fault diagnosis
particle swarm optimization
support vector machine
power transformer
parameter optimization
classification arithmetic
statistical learning theory
