摘要
文中提出一种基于支持向量机的变压器故障多层次诊断及定位模型。其基本思路是将变压器的油色谱信息和电气实验特征结合,再通过支持向量机对其进行学习分类,形成分层次、可靠、开放的变压器故障多层次诊断模型,并逐步对变压器的故障进行定位。充分利用支持向量机在解决小样本、非线性及高维模式识别问题等方面特有的优势,解决变压器故障信息存在的冗余、不确定、小样本等问题。实验证明,将支持向量机应用到变压器的故障诊断及定位中是合理可行的。
This paper presents a transfornler fault multi-level diagnosis and location model based on support vector machine. Firstly, the transformer oil chromatogram and characteristics of electrical experiments are combined and sent to support vector machine. Secondly, through machine learning and classification, the sub-level, reliable and open multi-level diagnosis model for transformer fault is obtained, by which the transformer fault is located step by step. This model fully utilize the advantages of support vector machine in solving the small sample, nonlinear and high dimensional pattern recognition and other aspects to solve the problems existing in transformer failure information about redundant, uncertainty, small sample size and other issues. The experiment showed that applying the support vector machine to transformer fault diagnosis and location is reasonable and feasible.
出处
《陕西电力》
2011年第8期46-49,57,共5页
Shanxi Electric Power
关键词
变压器
故障诊断
支持向量机
油中溶解气体分析
transformer
fault diagnosis
support vector machine
dissolved gas analysis