摘要
将粗糙集约简与支持向量机分类相结合,建立一个基于粗糙集和支持向量机的变压器故障诊断模型。该模型将油色谱数据诊断结果与电气试验数据相结合,通过粗糙集信息熵进行约简,建立了故障与信息的映射关系,再通过支持向量机分类器对其进行分类,使得变压器的故障分析到初步的部位。经实例分析和验证,该模型能对变压器进行初步定位,并有较高的正判率。
A model for the location of the transformer fault diagnosis is built based on rough sets and support vector machine. Firstly, the results of the oil data and the electrical experimental data are combined and reduced based on rough set theory, so the mapping of the faults and the information is established. Then the mapping is classified by the support vector machine classifier, so the rough faulty point of the transformer is diagnosed. Examples show that the proposed diagnosis model can get the rough faulty point of the transformer with a satisfactory accuracy.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2010年第18期80-83,共4页
Power System Protection and Control
