期刊论文

Analysis and Prediction of Transformer Overheating Fault Based on Chaotic Characteristics of Oil Chromatogram

中文

高压电器

1001-1609

2019

55

8

237-243

10.13296/j.1001-1609.hva.2019.08.034

本校为第一机构

在利用油色谱数据对电力变压器进行故障诊断时,为了改善因油色谱数据的波动及其非线性所带来的干扰,文中在利用油色谱分析过热故障时,引入了混沌理论对非线性系统进行分析。文中通过相空间重构建立油色谱数据混沌时间序列,并进行系统混沌特性判定,再将油色谱数据的波动以混沌特征量进行表征,比较分析过热故障特征气体在不同监测数据段中的混沌特征参量变化,从而提取油色谱非线性数据中所包含的状态信息。然后,基于混沌序列具有的短期预测性,利用最大Lyapunov指数预测法对油中气体体积分数的趋势进行预测。实例分析结果表明:过热故障特征气体的混沌特性参量能准确地进行过热故障的识别;在分析相应...

For the purpose of reducing the interference due to fluctuation and nonlinearity of oil chromatographic data in the fault diagnosis of transformer using oil chromatographic data, the chaos theory was introduced to analyze the nonlinear system. First,the chaotic time series of oil chromatographic data were established by phase space reconstruction, the chaotic characteristics of the system were determined, and the fluctuation of oil chromatographic data was represented by the chaotic characteristic parameters. The variations of the chaotic chara...